Henrik Mouritsen-Plebs and Politics in The Late Roman Republic-Cambridge University Press (2001)

This page intentionally left blank
`Plebs' and Politics in the Late Roman Republic analyses the political role
of the masses in a profoundly aristocratic society. Constitutionally the
populus Romanus wielded almost unlimited powers, controlling legisla-
tion and the election of ofcials, a fact which has inspired `democratic'
readings of the Roman republic. In this book a distinction is drawn
between the formal powers of the Roman people and the practical
realisation of these powers, or in other words between the Roman
people as a constitutional concept and the actual crowds which
represented them in public meetings and assemblies. The question is
approached from a quantitative as well as qualitative perspective,
asking how large these crowds were, and how their size affected their
social composition. Building on those investigations, the different types
of meetings and assemblies are analysed with a view to reconstructing
their practical functioning and locating them in their proper social
context. The result is a new picture of the place of the masses in the
running of the Roman state, which challenges the `democratic'
interpretation, and presents a society riven by social conicts and a
widening gap between rich and poor.
HENRI K MOURI TSEN is Lecturer in Ancient History at King's
College London. He is the author of Elections, Magistrates and
Municipal Elite: Studies in Pompeian Epigraphy (1988) and Italian
Unication: A Study in Ancient and Modern Historiography (1998).
Plebs and Politics in the
Late Roman Republic
Henrik Mouritsen
King's College London
iuniisuio n\ rui iiiss s\xoicari oi rui uxiviisir\ oi caxniioci
The Pitt Building, Trumpington Street, Cambridge, United Kingdom
caxniioci uxiviisir\ iiiss
The Edinburgh Building, Cambridge CB2 2RU, UK
40 West 20th Street, New York, NY 10011-4211, USA
477 Williamstown Road, Port Melbourne, VIC 3207, Australia
Ruiz de Alarcn 13, 28014 Madrid, Spain
Dock House, The Waterfront, Cape Town 8001, South Africa
http://www.cambridge.org
First published in printed format
ISBN 0-521-79100-6 hardback
ISBN 0-511-04114-4 eBook
Henrik Mouritsen 2004
2001
(netLibrary)
Contents
Acknowledgements page vi
1 Introduction: ideology and practice in Roman politics 1
2 The scale of late republican politics 18
3 The contio 38
4 Legislative assemblies 63
5 Elections 90
6 Plebs and politics 128
Appendix. The `Lex Licinia de sodalitatibus' 149
Bibliography 152
Index 162
v
Acknowledgements
This book started life as a research project sponsored by the Carlsberg
Foundation in Copenhagen. I am grateful for their generous support, as
well as for the friendly hospitality provided in that period by the
Department of History at University College London. I have also
enjoyed the help and advice of a number of colleagues and friends at
Cambridge, London and Leicester: Michael Crawford, Emmanuele
Curti, Emma Dench, Rebecca Flemming, Lin Foxhall, Peter Garnsey,
John Patterson and Jane Webster. Particular thanks are due to John
North, who has followed the project from its early stages and offered
useful criticism and encouragement throughout the process. Finally, I
would like to thank my friends and my family for their support and
interest in my work.
vi
1
1 Introduction: ideology and practice in
Roman politics
After years of relative neglect the role of the people in Roman politics
now attracts considerable interest among ancient historians. A wide
range of new interpretations has been presented, and at the root of this
interest lies a rediscovery of the fundamental paradox, which is the
Roman political system itself. On the one hand, the Roman people
wielded tremendous, almost unlimited powers. Their institutions con-
trolled legislation, declarations of war and the appointment of all state
ofcials; they were continuously consulted by their leaders and kept
informed through public meetings. On the other hand, Rome was also
an aristocratic society, where the elite controlled vast economic re-
sources and monopolised public ofce, political, military and religious.
The senate's inuence was overwhelming. It embodied all political
experience and religious authority in the Roman state, a position further
boosted by its successful leadership during the conquest of Italy and the
Mediterranean. This ambiguity has resulted in widely different assess-
ments of the nature of Roman politics, some of which can be traced all
the way back to ancient writers. Thus, according to Dio, 36.43.3,
Caesar `courted the good-will of the multitude, observing how much
stronger they were than the senate . . .' Sallust, on the other hand,
claimed that the affairs of the state were decided by `paucorum arbitrio',
because `plebis vis soluta atque dispersa in multitudine minus poterat',
Iug. 41.6. The question is therefore how these seemingly contradictory
systems coexisted; or in other words how much real power the senate's
ascendancy left the Roman people.
This book is thus an attempt to explore a familiar theme, and in this
introductory chapter the main issues and problems will be briey out-
lined. First, the recent surge in interest is placed in a wider historio-
graphic context, followed by a short discussion of the ancient attempts
at analysing the Roman `constitution' and the conceptual problems they
raise. The second part of this chapter looks at the question of ideology
and the relationship between political discourse and political practice in
ancient Rome.
2 Introduction: ideology and practice in Roman politics
Recent decades have seen a remarkable shift in academic interest
from the politics of the elite to the people's involvement in this process.
Radical new interpretations have been offered, also introducing the
notion of democracy, a concept which historians have traditionally been
careful to avoid in analyses of Roman politics. Historiographically this
development may be seen as part of a much broader sea-change in the
study of social structures and relations in republican Rome. Until the
last generation Roman historians paid little attention to the social strata
below the elite. Certainly in studies of Roman politics the masses were
largely ignored, with an overall emphasis placed on the structure of the
elite and its internal political manoeuvrings. Family alliances and poli-
tical groupings, ofce-holding and careers were the main foci of scho-
larly attention. This preoccupation was rmly rooted in the Rankean
tradition of the nineteenth century, which saw politics, diplomacy and
warfare as the proper subjects of historical research. Roman social and
economic history only developed slowly, and the study of Roman politics
long remained unaffected by these new trends. To some extent that may
have reected the broadly conservative outlook of most historians, for
whom Roman politics was often little more than a power game played
out between members of a few noble families. In accordance with this
concept, the existence of ideology in Roman politics was largely ruled
out. Even the emergence of `popular' politics in the second century,
which openly challenged the senate's supremacy and championed the
interests of the people, was following a hostile ancient tradition seen
as a barely disguised quest for personal power and prestige; little more
than an alternative way to advance one's career, using the comitia rather
than the curia as a stepping stone.
1
At the core of this approach thus lay
a somewhat cynical view of politics in general, bluntly expressed in
Syme's dictum that behind any political system, whatever it calls itself,
there always hides an oligarchy.
2
This detached and world-weary atti-
tude has characterised much ancient history, and has also led to a
general rejection of social issues as a signicant factor of Roman politics,
which some politicians might take up for other than selsh reasons.
The plebs itself was viewed with a certain disdain until recently most
ancient historians identied instinctively with the elite and readily
adopted its outlook and prejudices.
3
The lower classes were dismissed as
politically immature and entirely under the control of a few ruling
1
Meier (1965); (1980), which devoted a chapter to `die populare Methode'; Badian
(1972); Gruen (1974).
2
Syme (1939) 7.
3
Ibid., 100: `Debauched by demagogues and largess, the Roman people was ready for the
empire and the dispensation of bread and games.'
Introduction: ideology and practice in Roman politics 3
families. A nely woven network of clientela kept them in place and
reduced the popular assemblies to a political instrument in the hands of
senatorial factions.
4
The powers of the people were thus neutralised
through tight social control and the overriding aristocratic structure of
society in general.
This model of Roman politics has been challenged in recent decades.
The nature and extent of clientela have been questioned, and there is
now a growing consensus that its importance may have been over-
estimated.
5
For a number of reasons, to which we shall return in the
following chapters, it seems clear that the entire population cannot have
been individually tied to members of the elite. There is little evidence to
suggest the existence of a comprehensive network of social obligations,
linking top and bottom of society. That conclusion has important
implications for our picture of the people as a political agent; it opens up
the issue to new interpretations and forces the historian to reconsider
much received wisdom.
One response to this challenge has been the introduction of `democ-
racy' into the debate on Roman politics. This line has been most
forcefully advocated by Millar, whose recent book on the masses in late
republican politics represents a sustained attack on traditional posi-
tions.
6
Millar here presents the popular meetings and assemblies as
genuinely democratic institutions, which offered the Roman people a
crucial role in the political process and ample opportunity to make their
voice heard. There is an overall emphasis in Millar's work on the
centrality of these institutions to the workings of Roman politics. Far
from being a mere sideshow to the proceedings of the senate, they were
the focal point around which political life in Rome evolved. In public
speeches politicians of all persuasions addressed the assembled people
and put their case before them. In that respect the political system
approached what could broadly be termed a `democracy'. The opening
paradox of a seemingly cohabiting aristocracy and democracy has thus
been accepted as political reality, representing a genuine sharing of
power between the elite and the masses.
Viewed in a wider historiographical perspective, this rethinking of
Roman politics appears to be part of a more general development in the
4
Thus, the fundamental study by Gelzer (1912), later followed by e.g. Scullard (1973);
Meier (1980) 124; Bleicken (1974) 81; Gruen (1974) 365.
5
Brunt (1988); Wallace-Hadrill (1989); Pani (1997) 13240, 197.
6
Millar (1984); (1986); (1989); (1995a); (1995b); (1998), followed by Wiseman (1999).
Along the same lines also Lintott (1987); Purcell (1994); Laser (1997). This model has
sparked considerable debate: North (1990); Harris (1990); Gabba (1997); Molho,
Raaaub, Emlen eds. (1991); Burckhardt (1990); Jehne (1995); Ho lkeskamp (1995);
Flaig (1995); Pani (1997) 14055.
study of republican Rome, which also features the return of political
ideologies and a rehabilitation of the urban plebs. Along with the
`faction' model of elite politics, the narrow focus on the ofce-holding
class has largely been abandoned.
7
A growing number of historians also
accepts that Roman politicians might have been devoted to social
causes, which they pursued as part of a broader political agenda.
8
Our
perception of what Roman politics was about has thus been expanded; it
now seems to have dealt with real issues, which reached beyond the
internal power struggles of the elite.
9
Parallel to this development, the urban plebs has also been given a long
overdue rehabilitation. Few groups in history have suffered worse in the
hands of contemporary and later writers than the Roman plebs. In
antiquity the lower classes of the capital were vilied as parasites on the
state, fed by the public and overindulging in public entertainment. Until
recently the condescension of the Roman writers was perpetuated by
modern historians, who described the plebs in similar terms as a spoilt
and degenerate Lumpenproletariat. This attitude has nally given way to
a more balanced view.
10
Thus, historians have pointed out that the scale
and extent of public entertainment were far more limited than the
common stereotype would suggest.
11
Certainly under the republic it
was a diversion only a small minority of the population could enjoy, and
that just for a few days a year. Likewise public and private handouts
were insufcient to support a family. Rome's was therefore a working
population, which did not simply sponge off the state.
12
The result has
been a new picture of the lower classes in Rome, which suggests that far
from being overindulged, they suffered a precarious existence domi-
nated by frequent food-shortages, poor housing, high mortality and a
daily struggle for economic survival.
This rehabilitation of the plebeians, as we have seen, has been
accompanied by a wish to upgrade their importance as political agents.
Thus, the studies of, among others, Vanderbroeck, Yakobson and Millar
may be seen as further attempts to restore the dignity of the common
7
Against the faction model esp. Meier (1980), Brunt (1988).
8
Perelli (1982); Doblhofer (1990); Mackie (1992) esp. 6771.
9
Beard and Crawford (1985) 678, cf. the survey of modern research in Lintott (1994).
Brunt (1971b) is a ne example of this approach.
10
Important Yavetz (1958); Brunt (1966); (1980). Later contributions include e.g.
Ku hnert (1991); Will (1991); Prell (1997). This line has also been promoted by
Marxist historians, focused on class struggle in antiquity, e.g. Hahn (1975).
11
Balsdon (1969a); (1969b), who noted that `This army of unemployed idlers did not
exist', (1969a) 268.
12
Le Gall (1971); also Morel (1987) stressed Rome as a place of production, not just
consumption. More literature on the living conditions of the lower classes is listed in
ch. 6.
people and present them as serious players on the political scene.
13
They have presented an alternative to the traditional stereotype which
tended to see them as mere voting fodder, easily corruptible, devoid of
any serious political interests and therefore readily giving in to patronal
pressure; that is, if they did not happen to be carried away by the
rhetoric of `popular' demagogues. Millar has demonstrated in detail how
deeply involved the popular institutions were in the daily conduct of
politics, and sees the meetings and speeches to the crowd as indicative of
the independent and often decisive role played by the people, who were
carriers of distinct political interests, actively pursued in these fora. The
`democratic' interpretation may thus be seen as one strand of a general
basically sympathetic revaluation of the masses in history. As such it
draws moral authority and justication from this wider project but, as I
shall argue below, the social rehabilitation of the plebs as a `respectable'
working class may stand in the way of its political restoration. Material
necessity and political engagement might very well have been mutually
exclusive commitments. And what is interpreted as independence from
the elite may also be seen as separation from the political class and the
world it dominated, resulting in a general alienation of the plebs from
ofcial politics.
The picture of the Roman plebs as a responsible political agent, able to
provide qualied opposition to the senate, has found ancient support in
Polybius' description of the Roman political system as a `mixed' consti-
tution in which the popular assemblies represented the `democratic'
element.
14
This analysis is part of an ambitious attempt to explain the
wonder of Rome's conquest of the Mediterranean during the second
century. The focus is on the Roman `constitution', whose remarkable
strength is presented as one of the secrets behind Rome's success.
15
As a
Greek, writing for a Greek audience, Polybius' intellectual framework
was naturally that of traditional Greek political thinking. His main
analytical tool was the familiar model of the three constitutional arche-
types of monarchy, aristocracy and democracy, with their deviant forms
of tyranny, oligarchy and ochlocracy. The Roman political institutions
were tted into this particular scheme and each interpreted as repre-
senting one of these archetypes: the consuls represented the monarchical
aspect, the senate the aristocratic and the popular assemblies the demo-
cratic. Rome's was thus a `mixed' constitution which combined
13
Millar (1984); (1986); (1995a); (1998); Vanderbroeck (1987); Will (1991) 1ff.;
Yakobson (1992); (1999); Purcell (1994) 678; Pina Polo (1996) 12650.
14
Millar (1998) 24.
15
Polybius' discussion of the Roman `constitution' is presented in book six. See in general
Walbank (1972) 13056; Nicolet (1973); Nippel (1980) 14256.
elements of all three types, making it impossible to tell whether it was an
oligarchy, monarchy or democracy, 6.11.
16
According to Polybius Rome
had reached a balanced and therefore stable compromise, which
accommodated different forces in society and allowed her to transcend
the cycle of endemic constitutional change that was a feature of the
`pure' constitutional forms.
17
Thus, it was this `mixed' constitution that
gave Rome the strength and stability that enabled her to direct her
energy outwards towards military expansion.
Some scholars have argued that since Polybius had rst-hand experi-
ence of Roman politics, his description of the assemblies as a `demo-
cratic' element which counterbalanced the aristocracy should be taken
seriously, suggesting as it does that they in fact did function as effective
`democratic' institutions.
18
The terminological argument, however,
seems to underestimate the inuence exerted by Polybius' analytical
framework. Given the limitations of his conceptual `toolbox' it is
difcult to see which other term he could have used to describe the
popular assemblies. Polybius' application of his preconceived model to
the situation encountered in second-century Rome is quite schematic, at
times even crude. Strikingly so is his denition of the consulship as
`monarchical', which seems to ignore fundamental characteristics of this
ofce; its collegiality, the short-term tenure, and its appointment by the
comitia.
19
His approach to Roman politics is generally formalistic,
emphasising institutional structures rather than the practical workings
of politics. Polybius' analysis thus seems to owe more to Greek political
theory than to personal observation, and the terminology he uses may
therefore not be very helpful in determining the nature of popular
involvement in Roman politics.
Polybius' stress on the people as a counterbalance to the senate and
the consuls should be viewed in this perspective too. His aim was, as
noted, to explain Rome's extraordinary success, and his analysis there-
fore had to point to unique features which distinguished Rome from
other societies. The analysis followed conventional Greek patterns by
focusing on the Roman `constitution'; here Rome's superiority must be
demonstrated in terms that made sense to his audience. It is therefore
16
Generally, though, Polybius does recognise that the aristocracy carried the greater
weight in the Roman `constitution', cf. e.g. Pani (1997) 934.
17
Rome had not, however, completely transcended the ànacyclosis'. Polybius, 6.57,
predicted that eventually also the Roman system would decline under the inuence of
the imperial expansion.
18
Millar (1984) 3, followed by Walbank (1995), despite his more sceptical stance taken in
(1972) 155. Critical North (1990); Cornell (1991) 612; Flaig (1995) 88, 96.
19
This may be based on a Roman tradition, cf. Walbank (1995) 215. Still, such
subsequent rationalisation of its historical origins says nothing about its later
functioning.
hardly surprising to nd Rome described as an approximation of the
Aristotelian ideal of the `middle constitution', which mixes aristocratic
and democratic features. According to traditional Greek thinking only a
state that had neutralised the conict between these two was able to
reach the equilibrium which was seen as an essential feature of the ideal
constitution. The prominence accorded to the people in Polybius'
Roman `constitution' may therefore be seen as a product of the basic
theoretical propositions which informed his analysis.
20
It would have
served no purpose and made little explanatory sense to present Rome as
a society ruled by a small, mainly hereditary, ofce-holding elite. Such a
description would also, as we shall see, have been at variance with the
prevailing ideology and self-perception of the senatorial elite itself,
whose views Polybius had become deeply familiar with through his
friendship with Scipio Aemilianus.
The case of Polybius underlines the conceptual problems we are faced
with in trying to assess the scale of popular inuence on Roman politics.
The notion of `democracy' is problematic, coming to us as it does with
heavy historical and ideological baggage. Today `democracy' is hailed as
an almost universal telos and measure of human progress. As a political
principle it has been appropriated by virtually every regime in the world,
whatever its actual record. But not only is `democracy' one of the most
abused terms in the political vocabulary, modern notions of what in
practice constitutes a proper `democracy' are also very different from
those held in antiquity. Most crucially, the concept of political represen-
tation was unknown; any form of democracy was necessarily direct.
21
For those and other reasons `democracy' may not be very useful as an
analytical tool. Thus, the simple question, `Was Rome a democracy or
not?' by denition dees a straight answer. For while it may be possible
for autocratic regimes to eliminate at least temporarily most popular
inuence, probably no society can be totally democratic. The notion of
extending power equally to all citizens is in the nature of things very
difcult to realise in practice. `Democracy' would seem to represent an
ideal rather than an attainable goal. The question must be to what
extent a given system approaches this ideal, and there seems to be no
obvious way to proceed in such an investigation. A multiplicity of
criteria may be applied: the formal powers held by the people, the level
of direct popular participation, the equality of inuence and access to
public ofce, the accountability of ofcials to the people, the extent and
quality of public debates and consultation processes, the scope for
popular initiatives and policy-making, the inuence of outside bodies,
20
Walbank (1972) 155.
21
Cf. Finley (1983) 119.
such as pressure groups, lobbyists, and so on. Applied to republican
Rome each of them would lead to different conclusions, and none of
them would be able to tell us whether Rome was a `democracy' or not.
Instead of focusing narrowly on the concept of `democracy', it may be
more fruitful to follow the traditional Greek denition of the three
standard constitutions as those controlled by the one, the few and the
many. This is admittedly a very unsophisticated model, but the advan-
tage of `the rule of the many' over the abstract notion of `democracy' lies
in its implicit suggestion of a more practical approach to politics and
decision-making. As an ancient denition it reects the direct nature of
all political participation and draws attention to the actual scale of
politics. It shifts the overall emphasis away from the putative nature of a
constitution, its underlying principles or historical origins, onto the
practicalities of politics, thereby allowing us to ask more concrete and
therefore perhaps more answerable questions. This quantitative ap-
proach also has a crucial social dimension. For as Aristotle observed, the
few and the many are in fact the rich and the poor. `Rather, it is a
democracy when the free and the poor who are a majority have the
authority to rule, and an oligarchy when the rich and well born, who are
few, do', Pol. 1290
b
1820. The involvement of the poor would therefore
be a signicant indicator of the level of `democracy' in the Roman
republic. For `what does distinguish democracy and oligarchy from one
another is poverty and wealth', Aris. Pol. 1279
b
39.
22
The question is
therefore whether the popular institutions of Rome really did allow the
masses a say in the running of the state, giving them an opportunity to
pursue their own interests.
The practical denition of democracy as the `rule of the many' may
also help us to draw a clearer distinction between political ideal and
reality, that is, between the democratic potential and capability of a
political system and its actual functioning. There is no identity between
the level of participation and inuence a system theoretically or ideally
offers and the power the masses in reality hold.
This distinction between the ideals of popular political institutions
and their practical functioning has wide implication. We will have to
consider as a separate issue the ways in which the Roman res publica was
constructed ideologically. The popular institutions were part of a
complex system of values and ideals, which informed their procedures
and conventions. To understand the nature and functioning of these
institutions we must therefore look at the way the Romans thought
about themselves and their political system. What we nd is an almost
22
Ibid., 1011.
`democratic' discourse where the people appear as by far the most
important political body.
The centrality of the populus Romanus to the political debate is a
striking feature of late republican politics, where at rst glance it would
seem that the interests of the people were the primary concern of
anybody involved. The freedom of the Roman people, their libertas,
appears as the fundamental concept around which their institutions and
political practices were built.
23
This was the principle to which all public
orators, political theorists and historians paid tribute. The idea of the
civis Romanus as a free man in a free state is ever-present in the political
discourse of the late republic. The res publica was, as Cicero maintained,
really the res populi,
24
and the people were recognised as the foundation
of the Roman state and the ultimate source of political legitimacy.
Libertas was invoked by all Roman politicians whatever their views
and objectives. Political leaders otherwise ercely opposed to each other
were united in their common invocation of libertas as their guiding
principle. It may be less surprising to nd the Gracchi as champions of
the people's interests, e.g. Plut. Ti. Grac. 15.5, but their aristocratic foes
donned the same mantle and justied their actions as a defence of
libertas.
25
Later the Catilinarians conspired under the banner of liberty,
claiming to have the people's liberation from oppression and hardship at
heart.
26
However, after crushing the sedition, it was Cicero's turn to
present himself as the saviour of the res publica and its libertas.
27
More-
over, when he was exiled for his unlawful execution of the conspirators,
Clodius celebrated it as a vindication of the people's freedom and built a
shrine to Libertas on the site of Cicero's house.
28
On his return from
exile, however, Cicero claimed that not just his own person but also the
libertas of the Roman people had been restored.
29
Caesar went to war
against the established order not only to defend his own dignitas but also
to protect the libertas of the people.
30
Eventually he was killed by
disaffected senators who also acted in the name of the libertas populi
Romani.
31
Finally, when Augustus had established his personal
23
On libertas see Wirszubski (1950); Hellegouarc'h (1972) 54259; Stylow (1972); Fears
(1981) esp. 86975; Brunt (1988) 281350; Perelli (1990); De Martino (1989); Ritter
(1998).
24
E.g. Cic. Rep. 1.38; 3.43. Cf. Schoeld (1995) 6977.
25
Cic. Phil. 8.13; Brut. 212; Pis. 95; Planc. 70.
26
Sall. Cat. 20.14; 33.4; 58.8,11.
27
E.g. Cic. Sest. 123; Cat. 4.16.
28
Cic. Dom. 108, 110; Leg. 2.42; Plut. Cic. 33.1; Dio 38.17.6. Cf. Picard (1965); Berg
(1997).
29
Cic. Dom. 11011. Clodius is presented as an enemy of libertas in e.g. Pis. 15.
30
Caes. B. Civ. 1.22. Coins were also struck celebrating libertas, Crawford (1974)
no. 473. Both Caesar and Pompey claimed their allegiance to this ideal, Dio 41.57.2.
31
Dio 47.42.343.1. Cf. Crawford (1974) nos. 498502, 506.
monarchy, he too presented himself as the restorer of Rome's libertas
from the domination of a political faction.
32
The political discourse in Rome was dominated by an almost uni-
versal claim to be the true defender of the libertas populi Romani. It may
come as no surprise therefore to nd that there was little consensus
about the actual content of this notion. Freedom, then as now, was an
elastic concept, open to a variety of interpretations. There was in this
period broad agreement that the liberty of the Roman citizen involved
certain basic civil rights, among them the right to appeal, the ius
provocationis, which offered some protection against magisterial coer-
cion, and the right to tribunician assistance, auxilium. Libertas could in
this sense be used almost synonymously with civitas, citizenship. On the
other hand, the amount of political inuence entailed in the libertas of
the Roman citizen was understandably more controversial. On a
basic level libertas simply meant freedom from oppression, dominatio,
which again allowed for a range of interpretations, the most minimalist
merely identifying it with the absence of kingship. On this view libertas
was little more than res publica, a polity based on equality of citizenship
and a formalised system of power-sharing. At the other end of the scale,
however, libertas could be used to support calls for a much more
egalitarian distribution of power and wealth. Libertas in this form
represented the people's right to freedom, not just from kingship but
also from oligarchy. As such it was central to the agenda of the populares,
who applied a much wider interpretation than did traditional supporters
of senatorial authority, whose strategy was to limit the political impact of
libertas by reconciling it with the concept of dignitas.
33
For while libertas
was equal for all, dignitas was not; it reected your status in society.
34
Thus, while the historic right of Roman citizens to elect their own
leaders went unchallenged, it was at the same time argued that political
inuence had to reect the difference in dignitas. The two principles
were ingenuously blended in the comitia centuriata, which combined
equality of voting rights and disparity of inuence.
35
Cicero even
introduced a deviant form of libertas; the unrestrained rule of the people
which he dened not as liberty but as licentia, a disorderly state which
32
Res Gestae 1, cf. Syme (1939) 155; Ramage (1987) 6672. Augustus also presented
himself in his coinage as `vindex libertatis populi Romani', BMC 1 no. 691, Scheer
(1971).
33
Cic. Rep. 1.434. In 1.53 Cicero describes the ideal situation in which each citizen is
established in his proper station. For Cicero's concern about the gradus dignitatis see
e.g. Mur. 48.
34
This line was not new, cf. Cato maior, ORF
3
frg. 252: ìure, lege, libertate, re publica
communiter uti oportet; gloria atque honore, quomodo sibi quisque struxit'.
35
Cf. Cic. Rep. 2.3940. Cf. Di Gennaro (1993).
deed the auctoritas of the senate.
36
In the optimate discourse libertas
was therefore essentially the nobles' right to compete over public ofce
and inuence in an open contest with other members of their class. Its
antithesis was the regnum, the rule of one man. It was this particular
libertas that was lost with Caesar's dictatorship, and which Augustus
claimed to have formally restored.
Even from this brief sketch it is evident that libertas had no universally
recognised meaning, but could assume a variety of forms and be used
for very different purposes. We might draw the conclusion that it was an
empty slogan with no real meaning. Nothing, however, could be further
from the truth. The difculty we have in dening the exact content of
libertas is indicative not of its insignicance but of its fundamental
importance as the key political concept of the Roman republic. It was
central to the self-image of the Romans and at the heart of their political
identity. Libertas was the common ideal invoked by all Romans who
aspired to power, no matter what their political views and methods
might otherwise have been. Its highly varied usage illustrates the fact
that everyone involved in public life had to defend their position in
relation to this concept. All political acts and arguments must be
justied within the ideological framework of the liberty of the res publica
and the populus Romanus. And precisely because it was such a funda-
mental tenet of the identity of the Roman state, all political agents could
draw moral capital from it and exploit it for their own purposes.
The pervasiveness of libertas in the public discourse and its apparent
exibility are therefore indications of its overall importance; it was at the
core of the political system as it had developed in the early and middle
republic. When and how exactly that happened is not the issue here. But
it is important to note that the institutions of the late republic had been
shaped by an ideology which accepted that the entire citizen-body had a
legitimate claim to a share in the state. Early on, soldiers elected their
generals, and later the plebs appointed tribunes to protect its interests.
Legislation became the prerogative of the popular assemblies, to which
all citizens had access, and laws were passed to safeguard individuals
against magisterial force. The political institutions and practices re-
ected this basic idea of the identity of the state and the people, which
was further reinforced by a collective political mythology. The expulsion
of the kings had laid the foundation of the people's freedom,
37
which
36
Cic. Rep. 3.23: `si vero populus plurimum potest, omniaque eius arbitrio geruntur,
dicitur illa libertas, est vero licentia'. Also dened as ìmmoderata libertas', Cic. Q. Fr.
1.1.22; Flacc. 16, which allowed Cicero to attack democratic principles without
offending the universal ideal of libertas.
37
Livy 2.1; 2.15.3; 8.34.3; Cic. Parad. 12; Rep. 1.62.
had later been developed and extended in the `struggle of the orders'
and by subsequent legislation, cf. e.g. Cic. De or. 2.199. Libertas had
thus come to form a core element of the national identity of the
Romans. As Cicero declared, Ìt is to glory and to liberty we were born',
and this liberty was dened as a specically Roman value, which set
them apart from other nations, when he concluded that: Òther nations
can endure slavery: the assured possession of the Roman people is
liberty.'
38
This view was later echoed in Livy's comment that: ònly
those who took no thought for anything save liberty were worthy of
becoming Romans'.
39
The concept of libertas and the rights of the
populus were thus deeply ingrained in the way the Romans thought
about themselves and their society.
40
This value system structured the
public discourse, providing the framework within which all actions and
views had to be justied. But while the freedom of the Roman people
could not be questioned, this concept was also so vague and malleable
that it could be made to serve a multiplicity of purposes. In optimate
discourses senators could claim the ideal of libertas for themselves and
contrast it with licentia; even the term popularis was redened and
appropriated by politicians otherwise opposed to popular policies.
41
However, despite these artices it is important to remember that the
elite was not per se excluded from this particular construction of
romanitas.
The senate had a direct stake in this ideology, and a strong interest in
perpetuating it. To the senate there was no inherent contradiction
between its own claim to leadership and the libertas of the Roman
people. That was partly because of the limited interpretation applied
and the attempts to reconcile it to the principle of the gradus dignitatis,
which dissociated libertas from more radical democratic or egalitarian
principles. But it was also because the structure and value system of the
senatorial elite itself were compatible with this ideal. The senate's
authority was born out of the libertarian myth of the expulsion of the
kings, which had freed the people from the caprices of an individual
later the very essence of the senate's political creed. Subsequently the
nobility was created out of the historic compromise which admitted rich
plebeians to the ruling class. And even after the formation of the
38
Phil. 3.36: Àd decus et ad libertatem nati sumus', and 6.19: Àliae nationes servitutem
pati possunt, populi Romani est propria libertas.'
39
8.21.9: `. . . eos demum, qui nihil praeterquam de libertate cogitent, dignos esse qui
Romani ant'.
40
Libertas also became closely linked to imperium; the freedom of Roman citizens was at
the core of Rome's success abroad, cf. Sall. Cat. 7.3; Cic. Phil. 6.19. The Romans were,
in short, destined to rule others, not to be ruled themselves, cf. Cic. Phil. 4.1113.
41
Cic. Agr. 2.610, 102; cf. Seager (1972).
nobility, membership of the ordo senatorius did not become an inherited
privilege but was a status which had to be achieved personally in each
new generation. Generally that happened through ofce-holding; after
Sulla the quaestorate automatically gave access to the senate. The
political class was therefore able to claim a formal popular mandate,
which gave their position of power an element of meritocracy. In
principle and to some extent in practice the senate was open to
members of new families, and, as Hopkins and Burton have shown, a
large proportion of senatorial families did not maintain a permanent
seat in the Curia.
42
This meant that the senate was able to justify its pre-
eminent position within the same ideological framework of libertas which
provided the basis for the `popular' attacks on its authority. The senate
could draw no less legitimacy from the `people' and their institutions,
and was therefore more than willing to propagate the image of a
sovereign Roman people.
It is in this perspective that we must view the public nature of Roman
politics, which Millar has stressed as a `democratic' feature of the
republic. There was strong symbolic value attached to the appearance of
politicians before the people, addressing the assembled citizens and
gaining their formal consent. Such occasions were invested with a
broader meaning and functioned as public manifestations of the people's
libertas. Leaders of the state could publicly demonstrate their allegiance
to this fundamental principle, acknowledge the majesty of the Roman
people and claim their backing for themselves. Public meetings were
manifestations of a perhaps imaginary community between the
senatus and the populus Romanus. As such they may also have served a
wider purpose in maintaining social peace and stability by offering the
people a formal role in the political process, an aspect recently stressed
by a number of scholars.
43
Members of the senate actively promoted the ideological construction
of a res publica founded on libertas and popular sovereignty. On public
occasions leading senators and magistrates conrmed their devotion to
this ideal, and as a ritualised re-enactment of this principle the ofcial
meetings, contiones, developed their own internal logic. Contiones were
highly formalised occasions, on which the speakers by denition always
addressed the entire Roman people, no matter how small the actual
crowd may have been.
44
In this particular setting they automatically
42
Hopkins and Burton in Hopkins (1983) 31119.
43
Hopkins (1991) esp. 4925; Ho lkeskamp (1995); Flaig (1995); Bell (1997).
44
Cicero for example claimed that at the last contio of his consulship in 62 he had been
congratulated by `populus Romanus universus', Pis. 7, cf. Fam. 5.2.7; Rep. 1.7;
similarly the ùniversus populus Romanus' was present in the Forum, when the Lex
Gabinia was passed, Manil. 44.
represented the sovereign Roman people and embodied the ideal of
their libertas.
45
In assemblies and contiones `the people' were present as a
political concept, and it was this concept, rather than a random gath-
ering of individuals, which was addressed by the orators.
46
A speech to the people was therefore almost by denition `popular',
respectful of the Roman people and their majesty, attering and eu-
logistic in tone. The contio developed a set of rhetorical conventions,
which makes it very difcult to judge from the preserved speeches how it
functioned in practice. A good example is provided by Cicero's speech
against Rullus' land distribution scheme in 63. Here Cicero claimed
that Rullus had spoken disparagingly in the senate about the urban
plebeians who were the prospective beneciaries of his reform, and then
took the opportunity to show his own profound respect for this group,
Agr. 2.701.
47
We cannot, however, infer from these remarks in the
published version of Cicero's speech that he had actually addressed a
crowd of poor proletarians expecting to benet from Rullus' scheme. As
we shall see, that is rather unlikely in this instance, and Cicero probably
just followed the conventions of the contio. Irrespective of actual atten-
dance the orator always confronted the whole of the Roman people from
senators and knights down to the poorest citizen. In this case the speech
was also part of a concerted propaganda campaign against the bill, and
the shape of the arguments was naturally dictated by the prevailing
libertas ideology and focused on the best interests of the entire populus
Romanus. Since such debates were essentially contests in the most skilful
use of these concepts for the speaker's own advantage, public speeches
represent an intricate web of symbolic meanings and propagandist
effects, which may prevent any literal reading of their statements. They
constitute a discourse level which had its own logic, rules and conven-
tions. In principle, therefore, they may tell us very little about what
actually went on at contiones and assemblies.
The question is where this leaves our opening paradox and the
political inuence of the Roman people. Two different lines of interpret-
ation have developed in recent years. On the one hand, the `democratic'
version, proposed by Millar and others, suggests that the people's
ideological importance was simply a reection of their actual political
powers; politicians paid tribute to the people because they recognised
them as a dominant factor in the political life of the republic. On the
45
This aspect was symbolically underlined by the lictors bowing their fasces in respect for
the assembled populus, Cic. Rep. 1.62; 2.53; Livy 2.7.7.
46
Cf. Ho lkeskamp (1995) 38, who noted that the audience was the populus Romanus.
47
Cicero's hypocrisy is incidentally exposed in his letter to Atticus, 1.19.4, where he
himself describes the plebs as `sentinam urbis', `the dregs of the city'.
other hand, the people's involvement in politics might also be seen as
almost entirely symbolic; little more than a ritualised afrmation of
community, a view advocated by a number of scholars reacting to the
`democratic' school.
48
The former view would seem to underestimate the complexity of the
relationship between ideology and practice; a democratic discourse does
not necessarily reect a democratically functioning political system.
Ideologies are not mirrors of reality, but may and often do have a life
of their own, independent of the political practices at the time. In
modern times non-democratic systems routinely hide behind a screen of
populist rhetoric, obfuscation and pro forma elections. The separate
nature of publicly stated ideals and the ways power is exercised in
practice must be recognised; very often the former does not bear much
resemblance to the latter.
This criticism, however, does not invalidate `democratic' readings of
Roman politics, although it does question some of the assumptions on
which they have been based. A `symbolic' or purely ideological approach
to politics cannot disprove this model, since it operates on a different
level of abstraction. Recent attempts to dene contiones and assemblies
as `consensus rituals', merely serving to reinforce a symbolic community
between leaders and masses, have been both stimulating and innova-
tive.
49
But they are not a substitute for a practical analysis of the
functioning of the assemblies as decision-making bodies. For the point
is that these institutions had more than symbolic powers, and the wider
social functions they may have had do not affect their political substance
or potential impact. The application of a broader, sociological perspec-
tive, which recognises also the symbolic aspects of the political process,
may in itself tell us little about the political inuence of the masses. It
may explain why politically insignicant institutions and procedures
could still be socially important, but not why they were insignicant in
the rst place. We need to know exactly why they were mere rituals, and
why the people failed to exercise the powers which were formally theirs.
These questions have not been fully considered, and the model is
therefore not a proper response to the `democratic' challenge.
Thus, while the `symbolic' model, as it has been formulated, may
seem inadequate, the Roman republic clearly cannot be studied without
taking this dimension into account. It may help broaden our perspective
on the political process and raise awareness of the way in which the res
publica was constructed ideologically. It is paramount that we distinguish
the discourse on politics from the practice of politics. For while the
48
Most strongly argued in the volume edited by Jehne (1995).
49
See however the criticism in Pani (1997) 144.
populus may have held a crucial position in Roman political thinking and
its libertas have shaped its popular institutions, these very same popular
institutions were also based on the principle of direct participation.
Political inuence in Rome was never abstract but located very speci-
cally in time and space. It could be exercised only by citizens appearing
in a certain place at a certain moment. This primitive system created a
paradox; it meant that `the people', who formally represented the
primary source of political legitimacy in the Roman state, bore little
relation to the people who exercised these powers in the popular
institutions. There were two `peoples' in Roman politics: the ideal and
the actual. The people as a political concept were distinct from the
people as physical reality, and the direct nature of participation meant
that the two were effectively separated. There was a stark contrast
between the extensive powers of the populus Romanus as collective
political agent and the restrictive way in which this role was performed
in practice. The `people' may have been the foundation of the res
publica, but most citizens were in the nature of things prevented from
taking part in the political process. The structure of the popular institu-
tions thus takes us back to the simple distinction between the rule of the
many and the rule of the few. To approach the question of the degree of
`democracy' in Rome we will have to reduce politics to its most basic
components and consider the scale and capacity of the popular institu-
tions. The present impasse may be resolved by looking at those who
were present on these occasions, their numbers, composition, motiva-
tion and behavioural patterns.
The following chapters will therefore focus on precisely that aspect of
Roman politics which Millar and others have deliberately avoided.
While he concentrates on the political role of the crowds which `had for
whatever reason and in whatever numbers' convened for the meetings
and assemblies,
50
this study argues that the failure to distinguish
consistently between the `people' as a political concept and the `people'
as the sum of individuals making up the citizen-body is the main
weakness of the `democratic' interpretation.
51
While the former held a
central place in the political system, the latter were to a large extent
excluded by the workings of the assemblies. The formal openness of
these institutions to all citizens is in this perspective irrelevant. The level
of `democracy' must be judged by actual practice and not by an ideal or
50
Millar (1986) 4, cf. (1998) 45, 196, 212.
51
Millar is fully aware of the small numbers taking active part in Roman politics, e.g.
(1998) 37, but seems to see this as immaterial to the issue of Roman `democracy'.
Thus, he generally adopts the concepts and perspective of the ancient sources, speaking
in his opening paragraph of `the Roman people', àssembling in the Forum, listening to
orators there, and responding to them' (1998) 1.
theoretical entitlement to inuence. In Rome the people's institutions
may have been powerful, but that does not mean that the people as a
whole were. The seemingly contradictory statements by Dio and
Sallust, quoted at the start of this chapter, may reect this disparity, the
former referring to the extensive formal powers of the people, the latter
to the practical obstacles raised against their realisation.
This book is an attempt to explore the gap between the ideal and the
reality of Roman politics, between the populus Romanus and the crowds
which lled the Forum and the Saepta. The project faces obvious
difculties. Most scholars have considered it futile even to speculate on
the composition or motives of the crowds.
52
Nevertheless, the issue
seems too crucial to be ignored; it is, I believe, a key factor in under-
standing the political role of the popular institutions and the inuence of
the masses. One of the most serious problems is the patent lack of
interest in the issue shown by the ancient writers. No detailed or
objective discussion of this question exists. The sources either deal with
the people as a generic concept or give biased descriptions of their own
and their opponents' supporters, the latter evidently representing the
populus Romanus in a awed and deviant form.
53
The result has been that in most ancient texts the people appear as a
blank. In meetings and assemblies the populus was present as a constitu-
tional concept, and any political crowd was therefore automatically `the
people'. And as we saw, the politicians by denition always addressed
`the Roman people' when speaking at a contio. Likewise, the crowds
which passed laws or elected magistrates symbolically represented the
entire Roman people, and were described as such in our sources.
54
In order to assess the role of the àctual' people we will have to go
behind this ideological construction of the people as political agent, and
ask who in fact were the people appearing on these occasions. The rst
step is a quantitative assessment, establishing the overall scale of popular
participation. On that basis we will look in the next chapters at the ways
in which participation worked in the contiones, the legislative assemblies
and in the elective comitia. The aim is to present a picture of Roman
politics where the crowds are more than a generic concept, but have been
given individual features which reect the social diversity of the Roman
population as a whole.
52
E.g. Millar (1998) 137, 148.
53
Cicero, Sest. 126, claimed, for example, that the people supporting Ap. Claudius
Pulcher in 57 were not the real populus Romanus.
54
Thus, Cicero asserted that he had been elected consul by the entire Roman people,
Agr. 2.4, 7, despite the fact that he owed his consulship only to the vote of a relatively
small group of well-to-do citizens, registered in the rst two property classes.
2 The scale of late republican politics
The scale of political participation in the late republic may be ap-
proached from several angles. The physical setting for the assemblies
may provide an important indication as to the level of attendance
possible in these meetings. Likewise, there are scattered literary refer-
ences to the number of voters and the size of crowds; this evidence will
also be evaluated in this chapter.
The people convened in a number of locations in Rome during the
late republic: in the Comitium, the Forum Romanum, the Saepta, the
Circus Flaminius and on the Capitol.
1
The rst three venues will be
discussed in some detail in the following. The old Comitium, located in
the northwest corner of the Forum Romanum, was the traditional
meeting place for the tribal assembly and for the contiones, non-decision-
making meetings called by a magistrate. Little survives of the Comitium,
and although modern reconstructions differ substantially, there is broad
agreement on the overall scale of the site.
Coarelli reconstructed the Comitium as a circular area surrounded by
a cavea.
2
Exactly to the north was located the senate building, the Curia
Hostilia, which had access through the Comitium area. The entire
structure measured 46 metres in diameter, the central area itself being 30
metres across. However, large sections of the cavea, which was approxi-
mately 8 metres wide were not available for the crowds who, standing,
attended the assembly. The Curia would have intersected the cavea,
dividing it into two wings, or cornua as the sources describe them.
3
On
either side of the senate house were tribunals for the magistrates, in
addition to a considerable number of statues. To the south were located
the speaker's platform, the Rostra, and the platform reserved for foreign
embassies, the Graecostasis. The entrance, or entrances, to the Comi-
tium cannot be located, but they too would have reduced the space
available for the assembled crowds. Altogether at least a third of the
1
General summary in Thommen (1995).
2
Coarelli (1983) 11960; (1985) 1121, with comments by Krause (1981); critical
Vaathera (1993).
3
Cf. the reconstruction in Coarelli (1985) 120 g. 21.
18
The scale of late republican politics 19
cavea must have been taken up by other structures, the front of the
Curia, statues, tribunals, monuments, platforms and entrances. That
reduces the available space to about 1,300 square metres; tightly
packed, this might accommodate a maximum of 4,800 people.
4
This gure, however, may have borne no relation to the number of
citizens who could actually participate in the votes taken at the Comi-
tium. The system of corporate voting used by the Romans meant that
separate enclosures were needed for each of the thirty-ve tribal units
into which the citizenry was divided. During the voting the people
waited in their respective units, from which they were called forward to
cast their votes. We have evidence that rope partitions supported by
wooden poles were later used in the Forum.
5
Similar arrangements may
be envisaged for the Comitium. It is impossible to say anything about
the spatial organisation of these enclosures. However, since there was no
xed order of voting lots were drawn before each tribus was called
there would have to be enough room between the enclosures to provide
access to the `bridges' where the votes were delivered. Realistically
speaking, that would probably have reduced the effective waiting area by
at least a quarter. It is therefore difcult to see how more than 3,600
voters could have been accommodated in the Comitium.
An alternative reconstruction of the Comitium has recently been
suggested by Carafa, who envisages a roughly triangular structure,
separated from the Curia by the Via Sacra and by long stairs leading up
to the senate house, which was located on a low hill above the Comi-
tium. The space of the meeting area is calculated as 960 square metres,
allowing for a crowd of about 3,800.
6
How the corporate voting
procedures would have been carried out in such an irregular space is not
clear, and Carafa's model is also problematic in other respects.
7
4
Hansen (1995) 334 operates with 0.25 square metres per standing person, noting that in
Obwalden in Switzerland an assembly space of 1,000 square metres was tightly packed
when accommodating 4,000 voters.
5
Livy, 5.30.47, on an election in 393, which he locates in the `Forum', describes tribes
waiting to vote in separate units presumably surrounded by ropes, cf. Dion. Hal. 7.59.1.
Livy's account is probably inuenced by later electoral practices. Cic. Sest. 79 mentions
that the `manus Clodiana' used `fragmentis saeptorum et fustibus' as weapons. See
Taylor (1966a) 62. For Livy's use of `Forum' for `Comitium' in the early books see ead.
120 n.21, 130 nn.278.
6
Carafa (1998) 140. He puts the maximum capacity somewhat lower, assuming a crowd
density of only three persons per square metre.
7
For example the reconstruction leaves the cornua unexplained. Pliny, NH 34.26,
indicates that they were located in the immediate vicinity of the Curia. The ancient
sources generally insist on the close proximity between the Curia and the Comitium, e.g.
Cic. Q. Fr. 2.1.3; Livy 6.15.1; Varro, LL 5.155. Also the integrated Curia and circular
Comitium complexes found in the Latin colonies in Italy become very difcult to
explain on Carafa's model.
However, for this purpose it sufces to note that in terms of overall
capacity the two reconstructions are roughly in agreement.
These gures, of course, merely represents a maximum; the size of a
location tells us nothing about the actual attendance, which may never,
or rarely, have reached the full capacity of the venue. As we shall see,
there is no indication that when the Comitium was eventually aban-
doned for voting in 145 and for contiones in 122, it happened because of
growing difculties in accommodating the crowds within the limited
space available there.
In 145 the tribune C. Licinius Crassus proposed that the people be
given inuence over the co-optation of priests. The senate put up
strong opposition to this `popular' bill, and when it came to the vote
Licinius `was the rst to begin the practice of facing towards the Forum
in addressing the people', Cic. Lael. 96. Varro further explains that he
`was the rst to lead the people, for the hearings of laws, from the
Comitium to the voting area (or expanse) of the Forum', RR 1.2.9.
8
This move may not have been a practical solution to a spatial problem;
rather it seems to have been closely linked to his general stance of
deance towards the senate. As a symbolic act it makes good sense in
the architectural context of the Comitium, which on any interpretation
would have been visually dominated by the Curia building; Livy,
45.24.12, could even describe it as the `vestibulum' of the senate
house.
With this improvised gesture Licinius set an important precedent,
introducing the regular use of the Forum as a venue for legislative
comitia. It was obviously much more convenient to arrange the tribus in
the expanses of the Forum than in the Comitium. However, the initial
incentive behind the transfer had been ideological rather than practical.
It is therefore no indication that by 145 the crowds exceeded the
capacity of the Comitium.
The Forum evidently offered room for much larger crowds than did
8
`Primus instituit in forum versus agere cum populo', and `primus populum ad leges
accipiendas in septem [saepta?] iugera forensia e comitio eduxit'. For the alternative
reading of `septem' see Coarelli (1985) 1301. This passage undermines the claim by
Vaathera (1993) 114 n.95, 116, that the voters left the Comitium after the contio, were
lined up outside in the Forum, and then returned tributim to cast their vote in the
Comitium. The fact that Licinius `populum . . . e comitio eduxit' implies that people
had convened for the vote inside the Comitium and obviously were expected to stay
there. Moreover, even if the voters had left the Comitium, this space would still have set
the limits for the number of people able to attend the contio which preceded the vote.
The phrase used by the presiding magistrate ordering the contio to reform as a comitia/
concilium was `discedite', which does not indicate any spatial movement but the
formation into tribal units, cf. Asconius 71C. For the voting procedures in the Latin
colonies see Mouritsen, forthcoming.
the Comitium. In the rst century political activity was generally
concentrated around the Temple of Castor, where many meetings and
comitia took place. In front of the temple was a fairly large open space;
MacMullen estimates an area measuring 80 by 60 metres, able to hold a
maximum of 15,00020,000 people.
9
We may, however, obtain a more
precise picture of the capacity of this venue by looking at the practical
procedures for voting. The temple building itself was used for this
purpose, and in 58 Clodius blocked access to the temple by tearing
down the stairs. Cicero later claimed that Clodius had stored arms in
the temple, but it would also have prevented any vote from being taken
on the restoration of Cicero from his exile.
10
The Temple of Castor was
rebuilt at some point in the second century when, among other altera-
tions, the frontal stairs were transformed into a platform reached by a
ight of stairs on either side.
11
The actual delivery of the vote took place
on the pontes, raised wooden `bridges' designed to prevent outside
interference in the voting. A coin from 113112 shows the use of such
pontes in an election.
12
Taylor has made this reconstruction of the
arrangement on the Temple of Castor: `Since pons occurs in the plural in
descriptions of legislative assemblies, I suggest that the voters from each
tribe, called one by one into the precinct, marched to two pontes in two
lines on either side of the temple, and that the pontes over which they
walked to the voting basket were attached to the stairs on either side.'
13
This reconstruction is unsatisfactory, as it leaves unexplained how the
voters left the temple. Did they descend the same stairs by which they
had come up? in which case there would be no separation of voters
who had delivered their vote from those who had not. When looking at
the design of the temple from a voting perspective it seems logical to
assume that voters ascended by one set of stairs and descended after
9
MacMullen (1980) 456.
10
Cic. Pis. 23; Sest. 34, 85; Dom. 54, 110. Taylor (1966a) 41. Cerutti (1998) has argued
that the gradus in question was in fact a temporary wooden structure placed in front of
the temple and used for the voting procedures. The existence of such a structure,
though not impossible, is entirely hypothetical. Cerutti's theory is also difcult to
reconcile with Cicero's reference to the `gradus Castoris' and the `gradus eiusdem
templi', which clearly suggests that the stairs, destroyed by Clodius, were an integral
part of the temple. Moreover, it makes no sense to say that the destruction of the stairs
meant `sublato aditu' and the blocking of àditu atque ascensu', if only a temporary
wooden structure had been demolished, and the permanent lateral stairs on either side
of the podium were left intact.
11
Nielsen and Poulsen (1992) 806.
12
Crawford (1974) 3067, no. 292.
13
Taylor (1966a) 41, expanding her theory on page 45: ìt is possible that at the temple of
Castor and Divus Julius the cistae were on a lower step on which the pontes rested. In
that case the voter would have descended the steps immediately without going to the
tribunal.'
having voted by the other. The pontes would on this reconstruction
have been placed side by side on the central platform, which was about
7 metres deep. The space available here would probably not have
allowed for more than four pontes, considering the need for proper
separation and for ofcials to pass between them, handing out the
ballots to the voters.
14
On this basis some hypothetical calculations may be carried out with a
view to establishing an order of magnitude for the attendance possible in
this location. The basic framework is provided by the fact that legislation
was passed in a single day. To my knowledge no vote ever lasted longer
than one session. Some were interrupted or postponed for various
reasons, but that is irrelevant to our purpose.
15
Moreover, we never hear
of time being a critical element in the passing of laws; assemblies seem
to have tted easily into the time available.
Two more factors are needed: the speed of voting and the time for
counting and for the drawing of lots before each tribal vote. Both must
remain speculative.
16
Still, some cautious guesswork may be ventured.
Thus, concerning the speed of voting it would seem that hardly more
than four voters could have passed through a pons each minute. There
was probably only one voter at a time on the pons. The coin mentioned
above shows two people, one receiving the ballot, the other delivering it
at the urn. Still, that may be a visual compression of two stages in a
sequence. Consequently, a voter had to ascend the pons, receive the
ballot, scratch out one letter on the tablet and drop it into the cista,
14
This assumes that the pontes were placed next to each other parallel to the front of the
temple, where the presiding magistrate would have overseen the vote from a tribunal.
The pontes themselves would probably have been at least one metre wide, providing
space for the voter to pass the ballot box, which Licinius' coin indicates was placed on
the actual pons. The pontes would have been placed at some distance from each other,
allowing ofcials to pass between them and hand out ballots to voters standing on the
pontes. This arrangement may rst have been introduced with Marius' narrowing of the
pontes in 119, Cic. Leg. 3.38, which is the likely reference for Licinius' coin; before that
time the ofcials probably stood on the pontes. Marshall (1997) esp. 601, 678.
Cerutti (1998) has speculated that the vote took place on a landing of the wooden
stairs, which he hypothetically places in front of the temple. That seems most unlikely,
however, since the voting process had to be located in a formally inaugurated space, a
templum. He invokes the coin of Licinius in support of the theory, interpreting the
background as a tribunal with a subsellium. However, since the coin shows no stairs
leading up to the tribunal, the scene cannot be the one envisaged in front of the Temple
of Castor; also in the Saepta the presiding magistrate would have overseen the voting
proceedings from a tribunal.
15
In 133 at Ti. Gracchus' last (elective?) assembly he called off the voting, seeing that his
followers had not turned up in sufcient numbers. In 55 Cato obstructed the vote on
the Lex de provinciis by libustering, putting the voting off until the following day, Plut.
Cato Min. 43; Dio 39.34.
16
See also the attempts made by Staveley (1972) 18690.
before the next voter could receive his ballot.
17
The counting took place
after each tribus had voted, and the time taken up by that would have
varied according to the size and level of attendance of the tribe. Then
the result was announced and lots were drawn to decide which tribus was
the next one to be called forward. In general this procedure may have
taken as a minimum ve to ten minutes for each tribus.
18
On the basis of these estimates it would have taken a crowd of 10,000
citizens fteen hours to vote, provided that all tribes were called upon,
in addition to the time which the opening contio and the reading of the
law text would have lasted.
19
That probably brings us to a minimum
duration for the assembly of hardly less than seventeen hours. This does
not seem realistic. A more likely scenario would involve around 3,000
citizens who might have nished voting after six and a half hours. Most
often a majority would have been found before all tribes had voted.
Assuming that only two thirds or half the tribus were called upon, an
assembly of 10,000 citizens would have lasted nine and a half hours and
seven and a half hours, respectively, excluding the preceding contio.
These calculations, hypothetical as they are, merely serve to bring out
the extremely time-consuming nature of Roman voting; they suggest
that the voting facilities on the Temple of Castor could accommodate
many fewer voters than the open space in front of it. The maximum of
20,000, estimated by MacMullen, is far too high. Even a level of
attendance around 10,000 was possible only with a high degree of
unanimity among the tribus, many of whom would never have come to
vote. More realistically the average crowd would have fallen far below
this gure.
This result is in broad accordance with the few indications we have of
voter attendance in late republican assemblies. The sources tell us that
Ti. Gracchus had a following of 3,000 in his last dramatic assembly,
where it seems only his `hard-core' supporters had turned up.
20
The
reliability of such gures is of course compromised. However, a more
specic reference is given in Cicero's Pro Sestio 109, where we are told
17
For the procedure see most recently Luisi (1995).
18
An average of 7.5 minutes is used as the basis for the following calculations.
19
The full text of new laws was read out by a herald, cf. e.g. Asc. 58C, a procedure
which, given the length and detail of many bills, must have taken up considerable time.
Also the debates could be very time-consuming. Thus, in 55 Favonius was granted one
hour to argue the case against an extension of Caesar's command; Cato on the other
hand was given two hours, Dio. 39.34.2, cf. Millar (1998) 171.
20
Plut. Ti. Grac. 20.2. According to Sempronius Asellio, Ti. Gracchus at the time of his
death never went out with a following of fewer than 3,0004,000 men, Gell. 2.13.4.
This assembly may not have been legislative but elective, see Taylor (1963); (1966b);
Earl (1965); Hall (1998) 226. Plut. Sull. 8.2, also mentions that Sulpicius, tr. 88, had
with him a personal àrmy' of 3,000 men.
that some laws were passed in comitia in which some tribes were not
represented. In these cases ve voters were transferred from other tribus
in order to represent the missing ones.
21
It is entirely feasible, consid-
ering the maximum capacity of the Comitium and the Temple of
Castor, that some tribes were completely absent. Given the huge
difference in size of the tribus, an attendance of a few thousand voters
meant that the smaller ones would regularly have fallen so far below the
average that their units would have to be represented by members of
other tribus.
After the transfer of the legislative comitia the Comitium continued to
be used for contiones. The maximum capacity for meetings was, as we
have seen, higher than for the assemblies. But there is no reason to
assume a higher attendance in contiones than in the legislative comitia.
On the contrary, because of their political importance the comitia would
be expected to attract greater crowds than the contiones. The eventual
abandonment of the Comitium for this purpose was not the result of
capacity problems, either. Plutarch says that C. Gracchus, presumably
in 122, for the rst time turned his back on the Curia and addressed the
people gathered in the Forum, C. Grac. 5.3. The story has until recently
been dismissed as a repetition of Licinius' transfer in 145. That move,
however, only applied to the comitia. Both Varro and Cicero are explicit
in their references to the passing of laws, using the expressions àd leges
accipiendas' and àgere cum populo', the latter being a technical term
indicating the proposal of laws.
22
Archaeologically Plutarch's informa-
tion makes perfect sense. In Coarelli's reconstruction of the Comitium a
speaker standing on the Rostra would face the Forum when he turned
his back on the senate building. As in 145 this was a symbolic assertion
of the tribune's independence of the senate, denouncing the senate's
dominance over the people.
There is, in other words, no disagreement between Plutarch's story
and the references to Licinius' transfer of the comitia. Moreover, no
evidence suggests that contiones had been held in the Forum prior to
C. Gracchus' initiative. As mentioned above, the Temple of Castor had
been rebuilt at some stage in the second century 168 has been
tentatively suggested when a platform was constructed in front of the
pronaos. This was maintained when the temple was subsequently rebuilt
in 119 and later became a frequent location for contiones and comitia.
The original platform need not have been linked to any specic use for
contiones, however.
23
If in fact predating the transfer of the comitia, it
21
Sest. 109: Òmitto eas [leges], quae feruntur ita, vix ut quini, et ii ex aliena tribu, qui
suffragium ferant, reperiantur'. Cf. Tab. Heb. 334.
22
Cf. Gell. 13.16.3; Coarelli (1985) 158.
could have been used at a number of public occasions when audiences
were addressed, for example at funerals. In 142 Scipio spoke in front of
the Temple of Castor, `pro aede Castoris', but there is no reason to
assume this was a contio; it may have been linked to his censorial
functions that year.
In sum, the contiones were probably transferred to the Forum not
because of spatial constraints, but for purely ideological reasons. Pre-
sumably the space available in the Comitium was still sufcient for
contiones in the late second century. Indeed the maximum capacity of
about 4,0005,000 may rarely have been reached. Although specic
evidence for the level of attendance at meetings is lacking, we do have
some indications that very modest crowds may have convened on these
occasions.
24
Practical considerations would suggest natural limitations to the
numbers who could take part in a contio. In an era without loudspeakers
an orator's ability to address a mass audience was obviously limited.
Thus, Hansen has recently argued that the classic battle exhortations by
army generals may be a literary ction.
25
In practice no commander
could address an entire army lined up with their equipment. Moreover,
open spaces like the Forum or the area Capitolina were acoustically far
from ideal.
26
None of them were quiet secluded spaces. The Forum in
particular would have been quite noisy and unruly, making it even more
difcult for a speaker to make himself heard, cf. Asc. 41C. The Circus
Flaminius was also used for contiones in the late republic. This was
probably not, however, as the name might suggest, a built-up structure
similar to the Circus Maximus. As Wiseman has argued, it may origin-
ally have been just an open space, which on rare occasions was used as a
race course. In the late republic it became increasingly occupied by
public buildings.
27
The situation here would therefore not have been
different from that prevailing in the other venues. Here too there were
natural limits to the numbers who could attend a meeting, imposed by
23
According to Poulsen: `there is no evidence against it having been used in connection
with contiones' earlier than mid-second century, Nielsen and Poulsen (1992) 86. But
that overlooks the evidence of Plutarch, who explicitly states that C. Gracchus was the
rst to address the people in the Forum. See also Ulrichs (1994) 92.
24
Nowak (1973) 121, followed by Thommen (1989) 184, has calculated that Clodius
mobilised between 6,000 and 10,000 men on the basis of Cic. Dom. 119, where his
forces are compared to a consular army. That is obviously taking metaphors too
literally. In Dom. 80 Cicero refers to the men of one of Clodius' assistants as `Fidulii
centum', and in Sest. 59 the supporters of Clodius are given as `sescentas operas'.
These are, however, non-specic terms.
25
Hansen (1993).
26
Earlier the area Capitolina had been used for elective comitia tributa/concilia plebis, cf.
Taylor (1966a) 46, 132 n.38.
27
Wiseman (1974), followed by e.g. La Rocca (1995). Contra Coarelli (1997) 36374.
the practical problems of hearing the speaker.
28
For those reasons alone
the political crowds at contiones may never have grown much beyond the
capacity of the old Comitium.
Elections of magistrates in the late republic were held in the Saepta on
the Campus Martius. This location had always been the meeting place
for the comitia centuriata, which represented the Roman people as an
army and therefore had to convene outside the Pomerium. The comitia
tributa/concilium plebis, on the other hand, may only have been trans-
ferred there in the mid-second century perhaps not until the introduc-
tion of the secret ballot in 139, or shortly afterwards.
29
Previously it may
have taken place at the Comitium or on the Capitol (cf. Livy 34.53.2).
The possibility that the Comitium may have been used for elections in
the later second century suggests a very modest attendance in this
period, the location allowing for only a few thousand voters.
As the name suggests, the Saepta was basically an enclosure with
partitions for the voting units. It was also known as the òvile', referring
to the sub-divisions which resembled sheep pens. In the late republic
both the comitia centuriata and the comitia tributa used the same facilities,
which provided separate enclosures for the thirty-ve tribus; since the
third century also the voting of at least half of the 193 centuries had
been based on the tribal divisions.
30
In the period concerning us here
the Saepta was probably a temporary structure built of wood. It was not
until 54 that plans were rst made for a monumental rebuilding of the
Saepta. In a letter to Atticus, Cicero discusses current building projects
in Rome, noting that: Às for the Campus Martius, we are going to build
covered marble booths for the Assembly of the tribes and to surround
them with a high colonnade, a mile of it in all.'
31
The project, conceived
by Caesar, was eventually realised under Augustus, Dio 53.23.2.
28
On certain occasions more people might of course turn up than were able to hear the
speaker, e.g. the contio preceding the passing of the Lex Gabinia de imperio Cn. Pompei in
67, Cic. Manil. 44; in these instances, however, we are no longer dealing with a political
debate, intended to inform an audience of voters, but with a public demonstration of
popular support which may have been only indirectly linked to the political proceed-
ings. Thus, in the case of the Lex Gabinia we may doubt whether all the spectators
could have taken part in the actual voting.
29
Millar (1998) 25, 197, dates the transfer to the 140s, while Hall (1998) argues that it is
unlikely to predate the introduction of the secret ballot, and perhaps rst happened
when the practical difculties of the new, more time-consuming system became clear
and the practice of successive tribal voting had to be abandoned. Simultaneous voting
took up more space, thereby forcing a move to the much larger facilities of the Saepta.
The rst certain instance of a tribunal election held in the Saepta dates to 124, Plut.
C. Grac. 3.1.
30
Grieve (1985) has argued that only the rst class was divided into both centuriae and
tribus, but the fact that all classes used the same tribally organised voting facilities
would suggest that they were all structured along those lines.
31
4.16.8: Ìam in campo Martio saepta tributis comitiis marmorea sumus et tecta facturi
The location and shape of the Augustan Saepta have been identied
on the basis of fragments of the Severan map of Rome, the Forma
Urbis.
32
The relationship between this structure and the old Saepta is,
however, entirely hypothetical. Nonetheless, it may be argued that,
while its scale and details may differ, the overall shape was probably
maintained. The Augustan Saepta was a vast structure measuring 310
by 120 metres, orientated northsouth and located between the modern
Via del Seminario and Via del Gesu .
33
The fragments of the marble plan
show a large rectangular open space anked by two porticos, the
Porticus Meleagri and the Porticus Argonautici. Parts of the former,
rebuilt under Hadrian, are still extant to the east of the Pantheon.
34
The
north side, where the entrance lay, is known to have met the Aqua
Virgo, traces of which has been found in the Via del Seminario. South of
the Saepta a large hall, the Diribitorium, was built for the counting of
votes. A wall found beneath the Via del Gesu has been identied as the
south side of this building, thus indicating the total length of the
SaeptaDiribitorium complex.
35
In her book on the Roman assemblies Taylor, in collaboration with
Cozza, attempted a reconstruction of the inner space of the Saepta,
envisaging pens, measuring 260 by 2.5 metres, for each of the thirty-ve
tribes.
36
To the south they narrowed down as they led to a podium
where the actual voting took place. The corners of this platform have
been identied in some of the fragments. On the basis of this reconstruc-
tion Taylor calculated that the Saepta could hold a maximum of 70,000
voters.
37
MacMullen has since reduced this gure, partly by revising Taylor's
estimate of the crowd density, but most importantly by limiting the
eaque cingemus excelsa porticu, ut mille passuum conciatur'. Grieve (1985) 308
seems to suggest on this basis that the comitia centuriata did not use the Saepta. That is
beyond any doubt, however. The Saepta was the traditional location for the comitia
centuriata, and Cicero indicates that it continued to meet ìn Campo' throughout the
rst century BC, cf. Cic. Rab. Perd. 11; Cat. 1.11; Sull. 51. Alternatively, the comitia
centuriata would have moved to another, completely unknown location on the Campus
Martius, when the comitia tributa and the concilium plebis were transferred to their old
home in the late second century. Also the consular election of 45 (see below p. 301)
suggests the identity of the electoral venues in the rst century BC. Cicero's failure to
mention the comitia centuriata in his letter to Atticus is of no real consequence the
concilium plebis is also ignored here, and the specic reference to the comitia tributa may
be explained by the tribal voting enclosures, which provided the practical framework
for all the elective assemblies.
32
Carettoni et al. (1963) 97102; Rodriguez Almeida (1980) 129, pl. 27; Coarelli (1997).
33
The location is shown in Carettoni (1963) 98.
34
De Fine Licht (1966) 16370.
35
Muzzioli (1995).
36
Taylor (1966a) 54.
37
Followed by e.g. Demougin (1987) 310; Hopkins (1991) 495.
space available.
38
He pointed to an important inconsistency in the two
reconstructions of the fragments, presented in Cozza and Carrettoni's
publication of the Forma Urbis. In one plan the now lost fragment,
36a, `PORTIC[-]/ [-]AE[-]', is placed further to the north than in another
reconstruction shown in the same volume.
39
The rst solution, on
which Cozza and Taylor based their reconstruction, is unsatisfactory. It
not only entails a very odd spacing of `SAEPTA JULIA' and `PORTICUS
MELEAGRI'; the northern end of the building also becomes peculiar. The
lost fragment shows a wall with four openings, dividing the central area,
while the Porticus Meleagri continues to the north of this wall.
40
The
implication is that fragment 36a must be moved further south, thus
creating a forecourt between the entrance along the Via del Seminario
and the voting area south of the dividing wall. Thus, assuming a more
natural spacing of `SAEPTA JULIA' and `PORTICUS MELEAGRI', Mac-
Mullen estimated a voting area measuring 250 by 95 metres with `pens'
covering 200 by 75 metres, able to hold a maximum of 55,000 voters.
41
The signicance of this forecourt, however, does not lie simply in the
modication of the voting space which it entails. It is likely to have been
an integral part of the functional design of the building, and as such it may
provide a better indication of the crowds expected here than the voting
space itself. The principle of separating voters was essential to Roman
practice, cf. the rope partitions in the Forum. The procedures of the
comitia centuriata entailed a need for a second large enclosure next to the
voting precinct. The voting in the elective assemblies had been simulta-
neous at least since the introduction of the written ballot in 139. But in
the comitia centuriata this principle could apply only within the individual
property classes, which still voted successively.
42
It followed that only one
class at a time could occupy the central voting area; those who had not yet
38
MacMullen (1980) 454.
39
In tav. 31 the length of the open space is given as c. 250 metres and on page 98 as c. 285
metres. Rodriguez Almeida's plans (1980) pl. 27 and g. 36 are problematic too. To
make space for the Temple of Matidia, hypothetically located to the north of the
Saepta, the entire length of the Saepta is reduced from 310 to 285 metres, contrary to
all our evidence on the size of this building. This in turn creates inconsistencies in
Rodriguez Almeida's reconstructions of the slabs shown in gs. 34 and 36.
40
Rodriguez Almeida's reconstruction (1980) pl. 27, where the partition wall shown in
fragment 36a is identied as the facade, clearly demonstrates the problem of the
Porticus Meleagri's continuation to the north. Assuming that Caesar's project bore any
relation to the one eventually realised, this reconstruction is also called into question by
Cicero's description, which indicates that a portico would surround the whole voting
area, including the northern side.
41
MacMullen (1980) 454. On this reconstruction the word `PORTICUS' would have
crossed a section of the marble slabs, but that is not an uncommon feature in the
Forma Urbis.
42
Taylor (1966a) 96.
voted would have to wait in a separate enclosure before they could be
called forward. This separation of voters was crucial, in order to prevent
those who had already voted from rejoining those still waiting.
These practical measures to prevent fraud were essential since there
seems to have been no formal identity check or registration of voters.
43
This is suggested by the story about Marius' friend, Cassius Sabaco,
who was implicated when Marius faced accusations of bribery after the
praetorian election in 115, Plut. Mar. 5.34. One of Sabaco's slaves had
been sighted among the voters inside the partitions during the election.
Sabaco later explained that he had simply been thirsty and called for the
slave to bring him water; the slave had brought a cup into the enclosure,
and then had immediately left again. It is important to note that the
slave was not caught entering the pen; the case was based on a sighting
for which Sabaco was only later reproached. The implication is that no
identity check was conducted on entering the enclosures nor can this
have been the case at the actual voting. Otherwise the logic of the story
would be lost. If it was evident that the slave could not have voted
anyway, the accusation of fraud and Sabaco's reassurance that the slave
had immediately left the enclosure again, become meaningless.
The absence of any effective identity control made the separation of
voting groups essential. The forecourt to the Saepta would full this
purpose, thus solving the hitherto overlooked question of where the
43
Nicolet (1976) suggested that special voting tesserae were issued, but that theory has
been convincingly refuted by Virlouvet (1996).
Reconstruction of the Augustan Saepta. The central space is divided into two sections; a
forecourt for the waiting crowds and a voting area with `pens' for the electoral units.
other classes stayed while the rst class voted. In this respect the
Augustan Saepta probably perpetuated previous structures, incorpor-
ating an existing enclosure for the waiting crowd in the monumental
new scheme. It would also have been needed for the brief assembly
which preceded the vote.
44
It may thus provide a better indication of the
capacity of the Saepta than the pens themselves, which may have been
particularly spacious due to the special requirements of the voting
procedures. The maximum size of the forecourt allowed by the frag-
ments is 107 by 70 metres, not including the anking porticos, which
may have functioned as exits from the voting area. We cannot tell
whether this was a slightly expanded version of the previous design.
But considering the ambitious nature of Caesar's project he was in
competition with Pompey's theatre complex it is a possibility.
45
The
Augustan forecourt might hold around 30,000 voters, but a somewhat
lower maximum gure is probably more realistic. Otherwise the space
would be packed tight, making it very difcult to move around when
classes were called forward.
46
The Saepta Julia was clearly not supposed
to be crammed, and the court would also have been expected to hold
both senators and knights, waiting to deliver their vote.
Nicolet has drawn attention to the potential value of a rare reference
in Cicero, Fam. 7.30, to the length of time which the consular election
lasted in 45.
47
The circumstances were exceptional. The comitia tributa
had assembled to elect the new quaestors, when the presiding magistrate
decided to hold a consular election instead. Cicero tells how at the
second hour the tribal assembly had convened and by the seventh hour
the new consuls were announced. Consular elections were a formality at
this time, when only one consul was elected in the comitia, and he
44
For this contio see Taylor (1966a) 567, 136 n.64. Nicolet (1980) 250 has noted that in
Taylor's reconstruction the `pens' were too long, not leaving room for the contio. An
ancient formula recorded in Varro, LL 6.88, (cf. 6.94; Festus 100L) suggests that the
space for the contio was an enclosure especially designed for this purpose. The
magistrates called the contio using the phrase ìn licium omnes Quirites huc ad me', ìn
licium' referring to a precinct ritually roped in with strings, cf. ead. 56, 136 n.61, 156
n.41, Vaathera (1993). This primitive separation of the voting area from the meeting
precinct is likely to have been incorporated into the later architectural designs of the
Saepta complex. Virlouvet (1995) 1201 suggests that the contio was held north of the
Saepta.
45
Coarelli (1997) 15561 has shown that the orientation and overall scale of the Saepta
cannot have changed substantially since the third century. But that does not mean that
a minor expansion, for example to the north, may not have been possible in Caesar's
time.
46
Last-minute electioneering and political manoeuvring may also have taken place while
people were waiting, which would also have been difcult in a tightly packed crowd,
Val. Max. 4.5.3, Cic. ap. Asc. 85C, cf. Hall (1964) 28990.
47
`. . . hora secunda comitiis quaestoriis institutis . . . ille autem . . . centuriata habuit,
consulem hora septima renuntiavit', cf. Nicolet (1980) 291.
already had Caesar's recommendation. On this basis Nicolet estimates
an attendance of the rst two classes, sufcient to carry a majority of the
centuriae, of 16,800, assuming that the counting lasted one hour and
that the thirty-ve columns moved forward at the speed of two voters
per minute. This calculation, however, overestimates the attendance.
Firstly, the ve Roman hours recorded by Cicero were the equivalent of
only three hours and forty-ve minutes.
48
Secondly, the reorganisation
of the crowd following the magistrate's surprise decision to hold a
consular election would obviously have taken some time, perhaps as
much as twenty minutes. Thirdly, the process involved several time-
consuming reorganisations of the crowd. The separate vote of the
centuria praerogativa, the rst class, the equites and the second class
meant that new groups would have to be called forward and organised in
tribal columns, thus reducing the effective voting time, perhaps by as
much as twenty-ve minutes.
49
Finally, Nicolet overlooks the huge
difference in size of the tribus, which meant that the attendance in the
thirty-ve columns would have been highly uneven. The urban tribus
would probably have been virtually empty in the rst two classes.
50
Since the counting of one class and the voting of the following one could
begin only when the longest columns had nished, the effective voting
time for all the tribus would have been shorter. We have to deduct at
least a quarter of the time. By this calculation we reach a gure for the
attendance of the rst two classes at the quaestorian election in 45 of
around 6,000.
Another rare glimpse of the numbers involved in elections is given in
Cicero's speech for Plancius, who stood accused of electoral bribery.
Here we are told that the number of witnesses of the Voltinian tribus,
produced by Plancius' prosecutor, the defeated candidate Laterensis,
exceeded the votes he had received from this tribus.
51
The number of
witnesses in Roman court cases might be fairly large, but a delegation of
more than about ten witnesses would be highly unusual.
52
Laterensis
may have been particularly unsuccessful in the tribus Voltinia because of
the strong support Plancius enjoyed from Samnium. Still, Laterensis
was a serious contender, who seems to have expected an easy victory.
The fact that he only received single-digit returns from one of the large
tribus would suggest popular participation on a very modest scale.
48
Cf. Shackleton Bailey (196570) II, 434.
49
For the order of the voting see Cic. Phil. 2.82, cf. Nicolet (1980) 266.
50
Cf. Taylor (1966a) 95.
51
Planc. 54: `Nam quod questus es plures te testes habere de Voltinia quam quot in ea
tribu puncta tuleris . . .'
52
David (1992) 483 lists the known gures, and estimates an average delegation to have
numbered two to three persons.
The level of attendance was probably higher in elective than in
legislative comitia. Still, it would have remained far below the estimates
of Taylor and MacMullen. The attendance of a few hundreds from each
tribus seems more realistic. That would agree with the procedures used.
For the lack of formal identity check presupposes the possibility of
informal checks. The small numbers involved, probably with a strong
preponderance of the propertied classes, would have allowed the cura-
tores tribuum to know many voters by sight.
To sum up, it would seem that only a tiny proportion of the citizenry
ever took part in politics. Until 145 hardly more than 1 per cent of the
citizens could take part in legislation and perhaps also in the elections of
lower magistrates. This proportion must have fallen even further with
the expansion of the citizenry in the rst century, despite the change to a
larger venue. For the higher ofces the capacity of the Saepta could
hardly have accommodated more than 3 per cent of the approximately
910,000 voters registered in 70/69. Most of the citizens obviously lived
far from Rome, but even among those who had easy access to the
assemblies attendance would have been rare. The male adult citizens of
late republican Rome may have numbered about 200,000, a maximum
of 12 per cent of whom could have attended elections many fewer if
people came in from the countryside.
53
The level of attendance in
legislative assemblies and contiones would have been considerably lower.
We are, in short, dealing with a political process in which only a very
small section of the population ever took part. That in itself may surprise
few modern observers. Still, in the light of the extensive powers which
the assembly wielded, the limited degree of participation becomes some-
thing of a paradox. The disproportionately small scale of the democratic
institutions has been explained almost as a historical accident, the result
of Rome's extraordinary expansion from the middle republic onwards
combined with the prevailing self-perception as a traditional city-state.
54
Thus, while the citizen-body increased dramatically in this period, the
political institutions continued to work on the premise of a small face-
to-face society. Obviously, the growth of Rome created a huge gap
between the size of the citizen-body and the capacity of the political
institutions, but we may wonder whether that is the whole story. The
powers of the assembly may originally have been created in an environ-
ment of `democratic' struggle, but whatever the original character or
purpose of the popular institutions it is clear that by the later republic
no attempt was ever made to allow them to represent the political
53
For the urban population see Hopkins (1978) 968.
54
Bleicken (1975) 275, 281; MacMullen (1980). Along the same lines also Millar (1998)
21112.
interests of the people as a whole. The political `people', it seems, were
not identied with the actual masses, a point underlined by the absence
of any notion of numerical representation. Thus, there was no statutory
quorum prescribing minimum levels of attendance required to pass laws
or elect magistrates. Even the most far-reaching bills could be passed by
any number of citizens, as long as the correct procedures were followed.
The only regulation of this type we know of concerned the representa-
tion of all tribus. In case no tribules had turned up members of another
tribus could, as we have seen, be transferred to the empty unit.
The political indifference as to the numbers of citizens passing a law
or electing a magistrate is a striking feature of the Roman republic.
Unlike the senate, whose attendance was a constant concern for the
elite, we hardly ever hear of a well-attended assembly, `comitia fre-
quentia'.
55
In general politicians never claimed that their election or the
passing of a proposal had been effected by large crowds of people.
56
The
actual numbers were irrelevant to the authority of a law or election
result, none of which were ever challenged on the grounds of poor
attendance.
The case of Cicero's exile and restoration might seem to be an
exception. Cicero often contrasted the small crowds who had sent him
into exile with the overwhelming attendance at the assembly which
recalled him.
57
But the numbers, though constantly emphasised, are
secondary to the social composition of the crowds. Thus, the paucity of
Clodius' supporters implied in, for example, Sest. 53, seems less impor-
tant than their violent methods, servile origins and general depravity.
58
Likewise, Cicero's own following was not just numerous, it represented
the better part of the citizenry, and in particular that backbone of the
Roman state which Cicero identied as the local Italian elites. Cicero's
comparison between the crowds condemning him to exile and those
recalling him is therefore qualitative as much as quantitative.
Contrary to what might be expected, greater prestige or authority was
never drawn from the numbers attending an assembly. In elections it
was important to come out rst, but actual numbers are never referred
to. Large followings at public appearances were also a source of pride,
55
Cicero for example refers to the senate as `frequens' (in the sense of well-attended) no
less than twenty-three times, cf. Bonnefond-Coudry (1989) 41325, 420. A senatorial
quorum also applied in certain cases, ibid. 40113.
56
References to the size of comitia are extremely rare. Among the examples are Livy
45.36.6; Cic. Manil. 44; Fam. 8.14.1, (Caelius). Contiones might be described as large
or well-attended, e.g. Cic. Phil. 1.32; 4.1; 6.18; 14.5; 14.16; but here the implications
are different from comitia, see below.
57
Sest. 26; 53; 109; 131; Dom. 75; 90; Post red. sen. 18; 25; 28; Pis. 36; Fam. 1.9.13; Att.
4.1.4.
58
`lex erat lata vastato ac relicto foro et sicariis servisque tradito . . .', cf. Leg. 3.25, 45.
for example Cic. Att. 1.18.1, but that was distinct from the role of the
crowds as voters. Altogether this seems to reect a different concept of
legitimacy, one which was not associated with citizens turning up in
representative numbers to give their consent. The people's participation
in politics had a strong symbolic aspect to it, clearly brought out by the
rules laid down for tribal representation. Instead of an overall quorum
the attendance of all tribus was prescribed; as we saw, ve voters from
another tribus were to be allocated to vacant units. This emphasis on
symbolic, rather than real, representation is not far removed from the
practice prevailing in the comitia curiata, where, by the late republic, the
citizens did not turn up in person, but were represented by lictors, one
for each curia. It seems as if the people as political agent were detached
from the actual masses, with no clear perception of any direct link
between the two.
In practice the people's role was heavily circumscribed, which also
contributed to keeping participation down. That is particularly obvious
in the comitia centuriata, where the inuence of the votes was graduated
according to social standing. In the comitia tributa and the concilium
plebis the elite was stuck with a more democratic structure which gave all
votes equal weight. But here too some steps had been taken towards
limiting popular inuence, for example by placing freedmen in a few
urban tribes.
Still, in all types of assembly the classes outside the political elite
could in principle bring crucial inuence to bear on the Roman state.
The senate responded to this situation by trying to reduce the size of
crowds gathering for contiones and comitia. Thus, there are signs that
large-scale attendance at the assemblies early on was found undesirable
by the elite and actively discouraged.
Around 150 the Leges Aelia et Fua imposed an interval of three
market days, nundinae, before elections, where no comitia could be held.
In effect that prevented extra-urban voters, who might have come for
the annual elections, from exerting any inuence on legislation.
59
And
in 286 a law had already been passed prohibiting assemblies from
meeting on market days.
60
Again, the obvious implication was the
exclusion of citizens living outside Rome, who could no longer attend
the comitia when they came in for the markets. Even earlier attempts had
been made to reduce the attendance of extra-urban voters. Thus, the
59
Taylor (1962) 23 ignores this implication, seeing it simply as a practical measure.
Cicero, Post red. sen. 11, clearly saw it as a measure against popular mobilisation by the
tribunes.
60
Michels (1967) 105, De Ligt (1993) 112. According to Rutilius, cos. 105, the nundinae
had originally been instituted to allow rural voters to take part in politics, Macr. Sat.
1.16.34.
rst ambitus law, the Lex Poetilia from 358, banned campaigning by
magisterial candidates in market places and settlements, nundinas et
conciliabula, outside Rome, probably to curb the canvassing of novi
homines.
61
It thus reected the fundamental conict in Roman politics
between individual ambition, which aimed at the most effective mobili-
sation of personal supporters, and the overriding interest of the senate in
limiting popular participation.
In line with this policy no serious attempt was ever made in Rome to
accommodate the growing citizen-body in the assemblies. According to
Coarelli, the Comitium had been rebuilt twice between the early fourth
century and the mid-third century. In 338 the original square with 35.6
metre-long sides (1,267 square metres) had been expanded to 40 metres
(1600 square metres).
62
Later, around the middle of the third century, it
had been given its hellenistic circular form, which did not increase the
overall capacity; on the contrary, it probably meant a slight reduction.
63
In the same period, according to the census gures, the citizen-body was
almost doubled, as it grew from 165,000 in 340/39 to 297,797 in 252/1,
nally reaching 322,000 in 147/6, shortly before the Forum was rst
used for assemblies.
64
It seems clear also that in these early periods no
correlation was perceived between the number of citizens formally
entitled to attend the comitia and the crowds actually able to do so.
In the late republic there was a stark contradiction between the
formal powers of the assembly and the structural framework which
discouraged mass participation. To achieve large-scale political partici-
pation active promotion was needed, on an ideological as well as a
practical level. The constrast between the attitudes prevailing in Rome
and in classical Athens is instructive on this issue. The popular institu-
tions in Athens existed on a different scale. Even the fth-century Pnyx
I, measuring 2,400 square metres, was considerably larger than the
Roman Comitium; around 400 it was further expanded to 3,400
square metres.
65
The capacities of Pnyx I and II have been estimated at
6,000 and 8,500 respectively.
66
Much higher gures have been
61
Livy 7.15.123. Brunt (1988) 250 n.41 believes it to be anachronistic, in fact referring
to Gracchan times. Contra Ho lkeskamp (1987) 835; Cornell (1995) 469 n.33; Pani
(1997) 180.
62
Coarelli (1983) 11938.
63
Coarelli (1983) 1501 dates the circular Comitium between 263 and 252 BC. In
Carafa's reconstruction the space of the Comitium remained constant throughout the
entire period.
64
Brunt (1971a) 13.
65
Hansen (1991) 1289; Lotze (1995). The debate on the size and structure of the Pnyx
has recently been reopened in the light of new investigations, Forsen and Stanton
(1996).
66
Hansen (1996) 278.
suggested, but they are based on the unlikely assumption that the
Athenians stood rather than sat during the assemblies.
67
A quorum was
prescribed for certain types of assembly decisions, like ostracism and
grants of citizenship. This was set at 6,000. That gure, together with
the overall capacity of the Pnyx, must be seen against a demographic
background which was very different from the one obtaining in Rome.
The citizen population of fth-century Athens has been estimated at
40,00060,000, falling to 20,00030,000 in the fourth century.
68
To
maintain a high level of participation remuneration was therefore
introduced in 392, and it has been suggested that in the following
century as many as one fth of the citizens turned up for the ecclesia.
69
Evidently the rural population would have been relatively underrepre-
sented, correspondingly increasing the level of participation among the
urban citizens. Here the handouts also allowed the lower classes to turn
up in great strength.
70
In Rome, on the other hand, no effort was made to attract a
representative number of voters. Here the lower classes would have had
great practical difculty in participating on a regular basis. Political
activity in republican Rome was extremely time-consuming, and the
urban plebs clearly had other more pressing concerns. As noted in the
introduction, an important re-thinking has taken place in the study of
the urban population, which has emerged as very different from the
popular image of an idle proletariat, supported by the state.
71
Econom-
ically their lot was a precarious one. Free grain was not introduced until
58, and even then a family could not live entirely on the state dole.
72
In
order to survive, the Roman plebs had to earn its own living.
Political activity was therefore an economic sacrice for the working
population of Rome. In Sallust's description of the plebs' support for
Marius, this point is made explicitly: `Finally the commons were so
excited that all the craftsmen and farmers, whose prosperity and credit
depended on the labour of their own hands, left their work and attended
Marius, regarding their own necessities as less important than his
67
Stanton (1996).
68
Hansen (1991) 534, 904; Lotze (1995) 3978. In the fth century a considerable
proportion of the citizens were, moreover, away from Athens on military service.
69
Hansen (1991) 132.
70
Cf. Marble (1985).
71
See e.g. Purcell (1994).
72
The plebeians' reliance on their own labour, even after the introduction of the dole, is
underlined by Augustus' reorganisation of the grain distribution, which took into
account the fact that the recipients had pressing economic obligations and could not
take time off to queue for the handouts, Suet. Div. Aug. 40.2: `Populi recensum vicatim
egit, ac ne plebs frumentationum causa frequentius ab negotiis avocaretur, ter in
annum quaternum mensium tesseras dare destinavit', cf. Brunt (1980) 95.
success.'
73
In other words, the lower-class citizens who relied on their
own labour had abandoned their work and put Marius' career before
their own subsistence clearly an exceptional occurrence which de-
served special comment.
74
The implication is that working-class citizens
de facto would have been excluded from the political scene by the lack of
public remuneration, which in effect left the comitia in the hands of the
propertied classes for whom political activity did not entail any material
sacrices.
75
The small scale of the popular political institutions meant that they,
quite literally, represented the few rather than the many. Technically,
however, they remained open to a wide section of the population, which
held extensive formal powers in the Roman state. There was in other
words a marked contrast between the `democratic' potential of these
institutions and their limited format, which in reality excluded the
masses they formally represented. This peculiarity puts the focus on the
practice of politics in the late republic. Who turned up for political
events? What was their motivation? And how did they behave on these
occasions? Contiones and legislative comitia were the scene of what might
be called routine politics. In the next chapters these types of political
gathering will be discussed, before we turn to the focal point of Roman
politics, the annual elections of new magistrates.
73
Jug. 73.6: `Denique plebes sic adcensa uti opices agrestesque omnes, quorum res
desque in manibus sitae erant, relictis operibus frequentarent Marium et sua
necessaria post illius honorem ducerent.'
74
Cf. Cic. Cat. 4.16, noting that exceptionally even the poorest were present in the
large crowd which had assembled in the Forum awaiting the decision on the
Catilinarians: òmnis ingenuorum adest multitudo, etiam tenuissimorum'.
75
The introduction of the written ballot in Roman voting in the later second century
automatically excluded the illiterate masses of Rome, especially from the elective
comitia. Most likely, however, they would not have voted anyway. The fact that the
measure was regarded as a `democratic' reform would imply that it did not have
negative consequences for existing levels of popular participation. See Harris (1989)
16770, rightly critical of Best (1974).
3 The contio
A contio was a non-decision-making meeting called by a magistrate or
priest with ius contionandi.
1
Within these very broad terms the institution
appears to have been relatively exible. Thus, contiones could be purely
informative, communicating important news to the people, for example
military events, Livy 10.45.1; or emergencies, as happened in 184, when
the consul gave his famous address warning the people about the
Bacchanalian conspiracy, Livy 39.156.
2
Some contiones had an ofcial
function in presenting new legislation, which had to be put before the
people at least three market days before it could be voted on. These
legislative assemblies were themselves preceded by a special contio,
where the so-called suasio/dissuasio took place. This was a formal debate
on the proposal held immediately before the vote was taken in the
assembly. The two meetings were therefore intimately linked and their
attendance virtually identical.
3
These contiones must be distinguished
from those called ad hoc by a magistrate or priests, wishing to address
the people on any topic. It is the latter type which is our main concern in
this chapter.
The contio provided the only ofcial setting for political leaders to
meet the people, and the picture presented by the ancient sources is one
of lively civic events, which played a signicant part in the political life of
the republic. In recent years the contio has attracted considerable
scholarly interest, particularly among supporters of the `democratic'
model, for whom the institution represents an incontrovertible manifes-
tation of the people's crucial role in the running of the res publica. The
contio is identied as a focal point in the ongoing negotiation of power
between elite and populace, and the image of politicians addressing an
1
Thus the denitions in Gellius 13.16(15).3: `contionem habere est verba facere ad
populum sine ulla rogatione', and Festus, 34L: `contio signicat conventum non tamen
alium quam eum qui a magistratu vel a sacerdote publico per praeconem convocatur'.
Cf. Liebenam (1990), Taylor (1966a) 1533; Pina Polo (1995); (1996) 4852. For the
ius contionandi see Thommen (1989) 1719.
2
Generals could also deliver their reports at a contio, Livy 36.40.14; 45.40.9.
3
For the suasio/dissuasio see Mommsen (1887) III, 394.
38
The contio 39
assembled crowd of citizens, pleading their case and bringing all their
rhetorical skills to bear in an attempt to win popular support, might
indeed give the impression of a `democratic' process which was more
than a mere formality. In this chapter we shall therefore examine this
institution in greater detail, looking both at the issue of attendance and
at the wider function of these meetings in late republican politics.
The `democratic' reading of the contio assumes identity between
crowd and populus, that is between the people as a political body and the
individuals who formally represented it on these occasions. In Rome,
however, that equation is plainly untenable; as we have seen, only a tiny
proportion of the population could ever take part. The Roman citizen-
body could therefore produce an almost innite number of potential
audiences, which forces us to reconsider the composition of the crowds
that did attend the contiones. Our sources are rarely of much help on this
point. As noted above, they tend to refer to the audience in generic
terms as the populus or the multitudo. More detailed descriptions of the
composition of contiones are rare and, in the nature of things, often
coloured by political bias. This discussion, while taking into account
those few indications we do have, will therefore attempt a more struc-
tural approach, focusing on the practical aspects of conducting and
attending contiones in the late republic.
The fact that contiones were open to all citizens, but able to accom-
modate only a fraction of the urban population, puts the spotlight on the
small minority that did take part. Was the audience a miniature version
of the populus Romanus, representing the views and moods of the
populace as a whole? Did there, in other words, exist a particular plebs
contionalis, a mixed group of citizens regularly turning up for meetings
and assemblies, as has been suggested by some modern scholars?
Meier realised the implications of the small crowds involved in politics
and distinguished a small active section of the plebs, the plebs contionalis,
which he identied as a gathering of shopkeepers from around the
Forum.
4
In taking this minimalist stance Meier may have been inu-
enced by Mommsen, who believed the assemblies to have consisted of
merely a few hundred or thousand individuals collected from the back
alleys of the capital.
5
The idea of the plebs contionalis as a small
topographically concentrated crowd of traders raises practical problems.
4
Meier (1980) 114: `. . . kleinen Kramern aus dem Umkreis des Forums . . .' Although
these traders were few in number and socially unrepresentative, Meier also held that the
plebs contionalis rarely differed in its political views from the plebs urbana in general.
5
Mommsen (18545) II, 94, èine Masse vor allem, in welcher, von seltenen
Ausnahmefallen abgesehen, under dem Namen der Bu rgerschaft ein Paar hundert oder
tausend von den Gassen der Hauptstadt zufallig aufgegriffene Individuen handelten und
stimmten'.
40 The contio
For how would these people be able to attend often very frequent
meetings when they were supposed to earn a living for themselves?
Moreover, the theory fails to accommodate the changing venues for the
meetings. Thus, the contiones held on the Capitol or in the Circus
Flaminius would, on this line of argument, have attracted quite different
audiences.
Vanderbroeck, in his study of popular leadership in the late republic,
accepted the idea of a small group of people regularly participating in
meetings and assemblies. He described a plebs contionalis, which had a
`regularized behavioral pattern and anticipated collective behavior'.
6
And the membership of this plebs contionalis, with whom the `popular'
leaders `made a coalition', he identied as ìndependent freedmen', who
worked as shopkeepers and craftsmen.
7
An identication of the crowd as tabernarii and opices is in itself not
very illuminating; these professions remained the primary occupations
of the urban plebs.
8
The theory of a politically active section of the plebs
made up of independent freedmen is, on the other hand, based on a
combination of a few scattered references to freedmen in the contiones
(see p. 59 below) and a suggestion by Garnsey that some freedmen may
have been free from patronal obligations; some because they had
managed to buy off any remaining bonds, others due to their manumis-
sion ex testamento.
9
The existence of this category, likely as it may be,
does not alter the fact that most freedmen probably had fairly close ties
with their patron. Thus, the hypothesis does not lay claim to a general
reinterpretation of the relationship between freedmen and patrons.
Vanderbroeck's identication of politically independent plebeians with
freedmen is therefore paradoxical. Although some participation of
freedmen in meetings and assemblies may be plausible, an entirely
libertine plebs contionalis is not feasible. Inscribed as they were in only
four tribus, freedmen could not pass legislation. The implication is that
the otherwise stable plebs contionalis would have been substituted by a
completely different crowd when the contio was dissolved and the comitia
assembled to take the vote.
These attempts to dene a plebs contionalis illustrate the problematic
nature of the concept, which is basically a modern one. Cicero in
various contexts refers to `contionalis plebecula', `turba forensis' and
`populus contionarius', but a closer reading of these passages suggests
that no permanent plebs contionalis was implied.
The `contionalis plebecula', mentioned in Att. 1.16.11, appears in an
embittered invective against groups who had heckled Cicero in the past
6
Vanderbroeck (1987) 162.
7
Ibid., 161, 165.
8
Cf. Purcell (1994) 65973, and Treggiari (1980).
9
Garnsey (1981).
The contio 41
but now after his alignment with Pompey gave him a much more
favourable reception in public meetings. Cicero describes them as the
dregs of the city and a `starveling rabble' that `sucks the treasury dry'.
10
Clearly, therefore, the passage is not an objective description of a
permanent politicised section of the plebs; there is plenty of evidence to
suggest that boni often attended assemblies and public meetings. The
tone of the letter is generally exaggerated and the specic attack on
Pompey's supporters, invoking the recent restoration of the subsidised
grain dole, far too rhetorical to be of much use in determining who
attended contiones.
The `turba forensis' mentioned in De or. 1.118 is part of a common-
place attack on crowds cheering `popular' orators. He describes the
audiences as `barbaria' and the speakers as `vitiosissimi', most faulty.
11
Cicero is therefore not referring to a permanent `barbaric' crowd
occupying the Forum; the remark was directly linked to his dismissal of
`bad' speakers and has no wider application outside this specic context.
In Q. Fr. 2.3.4, Cicero lists a series of groups opposed to Pompey in
56, including the `contionarius populus'. As I shall argue below, they
may not represent a `politicised' section of the plebs, but rather those
citizens, often of higher social standing, who regularly frequented the
Forum and could be relied upon to turn up for a contio in support of the
senate.
Finally a passage in Plautus, Poen. 584, mentions `homines comi-
tiales'. But it is evident from the context that it refers to people hanging
out in the courts, not regular participants in legislative comitia or
contiones.
The concept of the plebs contionalis therefore has no ancient pedi-
gree,
12
and the question is whether it really helps us to better understand
the late republican contio. One feature which seems to emerge from the
sources is the seemingly erratic behaviour of the contional crowds,
whose sympathies appear to vacillate from staunchly optimate to ercely
popularis. This pattern of behaviour does not t easily into the picture of
a regular crowd attending the meetings. Vanderbroeck, however, identi-
ed an inherent volatility in the plebs contionalis, whose `loyalty . . . could
shift from one leader to another or even to the senate'.
13
He thus ends
up with a peculiar group of politically highly active, socially independent
freedmen whose continuous participation was not driven by any stable
loyalties or well-dened interests.
10
`sordem urbis et faecem', and ìlla contionalis hirudo aerari misera ac ieiuna plebecula'.
11
`. . . haec turba et barbaria forensis dat locum vel vitiosissimis oratoribus . . .'
12
Cf. Thommen (1989) 183 n.59.
13
42 The contio
The concept of a plebs contionalis creates more problems than it solves.
The main difculty involved concerns the ability of its members to
attend on a regular basis and their motivation for doing so. For why
would a small section of the plebs regularly turn up for political meetings
in which it had no voice apart from cheering and jeering and which
took no decisions? Precisely because the overall scale appears to have
been so modest, participation cannot have been part of their daily
routines or embedded in any political culture among the plebs. The plebs
contionalis would therefore have been a highly specialised sub-group of
politically minded citizens, set off from a largely inactive or depoliti-
cised majority of plebeians. Viewed in this perspective the theory
clearly has more than a touch of anachronism; such a politically
concerned public outside the ruling circles did not emerge until the
rise of the bourgeoisie in the early modern period.
The social background of the participants raises yet another problem:
how was it economically feasible for a working-class plebs contionalis to
invest so much time in politics? As we have seen, the image of the
plebeians as mere idlers living off the state dole and private benefactions
has now been questioned. It seems that the large majority of the Roman
population relied on their own work for their economic survival.
Political participation, on the other hand, could be very time-con-
suming. There were few restrictions on which days could be used for
contiones. And although direct evidence is lacking, it seems that contiones
might be called with great frequency. Thus, in Sest. 39 and 42 we are
told that Clodius held daily contiones against Cicero.
14
Meetings were
often called a day in advance; formally it happened by praecones going
round the vici announcing the contio. But it remains questionable how
many tabernarii and opices would have been able to leave their shops/
workshops at such short notice. We know little about the duration of
contiones. Presumably it might vary considerably. Some featured a
number of speakers and would probably have lasted several hours, to
which should be added also the time it would have taken to get to the
venue from the residential areas of the city. Viewed in this perspective
regular participation would not have been a realistic option for most
plebeians. Tabernarii may on occasion have been able to close their
shops, but for others, for example day-labourers, that was not an
option. Given the economic sacrices involved, the idea of some
members of the plebs routinely leaving their work in order to listen to
speeches by the elite seems less plausible.
14
Cf. Cic. Mil. 12; Cluent. 93; 103 (Quinctius); Marc. 27; Verr. 2.3.223 (Cotta). Asc.
51C, who refers to `cotidianis contionibus', held by the tribunes after Clodius' death in
52. In Brutus 3056 Cicero mentions frequent contiones held in 91.
The contio 43
As a working hypothesis the plebs contionalis is not very useful. There
is no rm evidence to suggest the existence of a group of `politicised'
plebeians. There are, moreover, numerous instances from the late
republic which cannot possibly be tted into a model of a more or less
permanent plebs contionalis. However, before examining this evidence, it
may be worth pursuing further the social implications of this discussion.
For it follows logically from these considerations that the political crowd
in Rome generally would represent the propertied classes rather than
the working population. A system which offers no incentive for popular
participation is naturally left to those with time and resources, both
essential requirements for regular political participation at Rome. It
would therefore not be surprising if the political populus addressed by
the politicians turned out to be socially far superior to the mass of urban
plebeians, and there are in fact indications that under normal circum-
stances public life was indeed dominated by the propertied classes.
We have already seen how Sallust noted the unusual turnout of lower-
class citizens in support of Marius, and Diodorus describes Octavius'
followers in 133 as `not just recently assembled', but `the most politically
alert and the well-to-do segments of the populace', 34/35.6.2.
15
One
notes that the àctive' element was also the wealthy one, and its support
for the senate against Ti. Gracchus suggests where the natural interests
of this group lay. For the point is that outside the ofce-holding class
there were also people with considerable economic resources, who did
not rely on their own labour but could dispose freely of their time. In
this vital respect they represented the natural political crowd in repub-
lican Rome. They could command the attention of the ofce-holding
class, and expect to be formally consulted on political issues. Given their
social standing, they would presumably also have been well integrated
into political circles and better informed about current affairs than was
the mass of urban citizens.
The setting of contiones in the Forum may also have reected and
further encouraged regular participation by this group. The forum has
sometimes been described as a popular space where Romans from all
walks of life would come together, a picture heavily reliant on Plautus'
vivid sketch of social types in the Forum, Curc. 45582.
16
Here comic
exaggeration is evident, however, and by the late republic the social
reality of the Forum may have been less diverse than Plautus' ctional
account would suggest. The sources give several hints that in this period
15
àlla to praktikotaton tou demou kai tois biois karpimou'. Diodorus' hostility towards
the Gracchi is well known; still, that does not explain his description of their
opponents.
16
E.g. Rouland (1981) 113.
44 The contio
the Forum was dominated by men of substance. In the Commentariolum
Cicero is advised to secure the centuriae through senators, knights and
active and inuential people from all other ranks, with the addition that:
`Many energetic city folk, many inuential and active freedmen are
about in the Forum.'
17
The forum crowd is here set apart from the
urban plebs at large, briey mentioned in the following paragraph, and
their description as `gratiosi' to be approached personally by the candi-
date suggests that they were men of means and inuence. Cicero himself
also described the Forum as a dignied space `full of the best men and
citizens', explicitly contrasted with the popular Greek agora and the
unruly assemblies of the East.
18
C. Gracchus warned against following Capua's example and ex-
cluding the plebs from her political centre, but not only was he then
pursuing a very specic ideological point,
19
there is also evidence that
the Forum was gradually `cleaned up' during the later republic and
turned into a monumentalised formal space.
20
Sordid trades were
removed from the Forum, and forced into the side-streets or other parts
of the city. What remained of commerce were bankers and luxury shops,
catering for a wealthy clientele. In this period, therefore, we nd little
trace of the colourful social mix presented by Plautus. The Roman
forum emerges as a dignied place for public affairs, where respectable
citizens would come together to attend court cases, socialise, shop,
conduct nancial transactions, accompany important gures or friends
running for high ofce.
The political stance taken by the Forum crowd points in the same
direction. Thus, Cicero refers to the senate's champion Curio, who
received `hearty rounds of applause, a most attering amount of general
salutation in the Forum, and a great many other signs of good will from
the boni', Att. 2.18.1.
21
Their senatorial stance and opposition to Caesar
are further underlined by the eagerness with which the people in the
17
Comm. pet. 29: Èt primum, id quod ante oculos est, senatores equitesque Romanos
ceterorum [ordinum] omnium navos homines et gratiosos complectere. Multi homines
urbani industrii, multi libertini in foro gratiosi navique versantur.' In Fam. 5.15.2
Cicero notes that: `forum commune sit', but he does so in a letter to Lucceius,
suggesting that this was indeed the place where he would go to meet fellow senators.
18
Cic. Flacc. 57: `plenum optimorum virorum et civium'.
19
Val. Max. 9.5 ext. 4. Purcell (1995) takes Gracchus' ideological point literally and
distinguishes a `plebeian' and a `noble' forum in Rome. However, the evidence
adduced for the former remains very limited and ambiguous.
20
A process traced in Morel (1987) 1357, who notes the gradual transformation of the
Forum, which left only `respectable' forms of commerce, cf. 145. Varro, De vita pop.
Rom. 2 fr.72, for example, mentions the removal of the butchers' shops from the
Forum.
21
`consalutatio forensis perhonorica, signa praeterea benevolentiae permulta a bonis
impertiuntur'.
The contio 45
Forum gathered to read Bibulus' decrees, Cic. Att. 2.21.4.
22
There are
also other examples of the Forum crowd rallying to support the senate
against `popular' opponents (below pp. 512). The `contionarius
populus', mentioned by Cicero in Q. Fr. 2.3.4, becomes interesting in
this context. They feature alongside the nobilitas, the senate and the
iuventus (probably of good family, cf. Comm. pet. 6) in a list of groups
opposed to Pompey in 56. These meeting-going citizens are clearly
distinguished from the popular supporters of Clodius, who are men-
tioned separately in the same paragraph, and their association with the
elite might suggest that they were respectable boni frequenting the
Forum on a regular basis.
The notion of the Forum as a meeting place lled by a cross-section
of the urban population may therefore be too idealised; more realistically
the Forum belonged to the world of the elite rather than the populace in
general. The small crowds gathered for the contiones may typically have
been men of substance; attending the Forum and the public meetings
held there probably formed part of the daily routines associated with the
leisured lifestyle of a Roman gentleman.
Outsiders were not excluded from these occasions that would have
breached the basic principles underpinning the contio as an institution.
Anyone interested in a political issue could turn up, but it remains an
open question how many members of the lower classes would have done
so.
23
Apart from the material disincentive, the nature of the issues
brought up in public meetings would also have militated against regular
plebeian participation. In general `Roman politics' was rmly rooted in
the world of the elite, and many issues were marginal to the lives of the
masses. We may wonder, for example, how many shopkeepers would
rush to the Forum to listen to speeches on aspects of foreign policy or
regulations for ofce holding. How would members of the lower plebs
acquire an interest in topics of this type, or gain any information about
them?
On this interpretation the existence of the contio does not in itself
represent a challenge to traditional readings of Roman politics as
fundamentally aristocratic in nature. As argued above, the contio can be
seen as a formal consultation of the people for whom politics mattered
22
Cf. the anti-triumviral feelings of the boni in the theatre, Cic. Att. 2.19.23 and 4. The
fact that the Forum crowd was able to read Bibulus' decrees also suggests that they
belonged to the literate classes.
23
Nippel (1995) 47 seems to suggest a model of issue-driven participation, where people
turned up for contiones on causes close to their heart or affecting them personally. Given
the fact that it was the elite which formulated the political agenda in Rome, the
implication of such a model is a political crowd heavily dominated by this very same
elite.
46 The contio
and who mattered to the politicians. As such it may simply have
extended the political process to a broader section of the elite outside
the active political class of ofce-holders and senators, while at the same
time paying tribute to the venerable ideals of citizenship and libertas.
The fact that political proceedings are public does not in itself make
them `democratic'. That depends entirely on who attends them and, as
we shall see, on the relationship between the meetings and the political
decision-making.
The `democratic' model presupposes a direct link between public
debates and the political actions taken by the sovereign populus. But the
connection between argumentation and decision-making is undermined
by the scale of the meetings compared to the citizen population as a
whole. For apart from the pre-comitial debates, contiones did not address
the electorate in preparation for a forthcoming vote. Given the logistics
of the meetings, the crowd turning up for the nal vote might in
principle be completely different from the one which had attended the
preceding contiones. This situation raises the question of why these
meetings were held at all, or, in other words, what their function was in
late republican politics.
The most striking aspect of the Roman contio is the fact that, unlike
the Athenian assemblies, it was not open for everyone to put forward
their views. The contio remained under the control of the presiding
magistrate, and only with his permission could others be allowed to
address the meeting.
24
Thus, the possibility of excluding debate is
illustrated by Cicero's description in Pro Cluentio of a contio in which no
dissenting voices were allowed: `The case was taken up at public meet-
ings; and though it had never been heard, the same view of it was taken
by the populace. No one had a chance to denounce that view; no one in
fact exerted himself to urge the opposite view.'
25
The presiding magis-
trate was, in other words, free to exclude any opposition from his
meeting.
The contio was not designed as a forum for open debate, not even
between members of the political class. And certainly it did not provide
a free exchange of arguments informing the sovereign Roman people.
The contio was essentially an ofcial platform for politicians to present
themselves and their views to small, presumably inuential audiences.
As Gellius dened it, a contio was simply `verba facere ad populum', and
24
Cf. Ho lkeskamp (1995) 35.
25
Cluent. 130: Ìactata res erat in contione . . . incognita causa probatum erat illud
multitudini; nemini licitum est contra dicere, nemo denique ut defenderet contrariam
partem laborabat.'
The contio 47
its remoteness from the liberal ideal of an open democratic discourse
becomes even more apparent in the late republic, when traditional
political structures collapsed under the pressures of increased elite
competition and the emerging conict between senate and populares.
For in this increasingly anarchic period it becomes a real issue whether
such a `democratic' institution, if intended, would have been possible.
In practical terms the question is whether a Roman orator would have
been able to address a crowd which was hostile or even politically split.
The simple fact that orators relied entirely on the power of their own
voices, often in acoustically difcult locations, made them extremely
vulnerable to noisy or unruly crowds. Many instances are on record
which suggest that simply to be heard, orators to a large extent
depended on the good will of their audience.
Hissing and shouting were common ways of showing disapproval on
public occasions. Cicero notes that: `Those ``populares'' have taught
even decent people to hiss.'
26
But the problem was not new; already in
169 Livy, 43.16.8, reports a contio where the heckling of one particular
speaker became so intense that he had to call the herald to bring the
meeting to order.
27
At the contio preceding the passing of the Lex
Gabinia, which granted the pirate command to Pompey, his senatorial
opponents were harassed by the crowd; only when Catulus spoke was it
`quiet for some time', Plut. Pomp. 25. Pompey was shouted down when
he tried to defend Milo in 56, Cic. Fam. 1.56.1; Q. Fr. 2.3.2; Plut.
Pomp. 48.7.65, Cic. Fam. 1.5b.1; Plut. Pomp. 48.7; Cic. Q. Fr. 2.3.2.
Cicero himself was jeered at the political trial of Rabirius, Cic. Rab.
perd. 18. According to Appian B. Civ., 2.131, Lepidus could not make
himself heard against the shouting of a hired crowd. At the trial of Milo
soldiers were called upon to secure a silent hearing, Cic. Milo. 3;
Asconius explains that: `This subdued the Clodians, and for two days
they allowed the witnesses to be heard in silence.'
28
In 57 the senate
made the consul call a contio, which explicitly excluded the followers of
Clodius, so that `no ears should be affronted by the voice of any hireling
or scoundrel raised in bitterness or enmity against the boni'.
29
The
haughty demand for silence made by P. Cornelius Scipio Nasica, consul
26
Att. 2.19.2: `Populares isti iam modestos homines sibilare docuerunt.' The fact that this
type of behaviour was not limited to the `lower classes' is underlined also by Cic. Att.
2.18.1, describing how the boni pursued Fuus `with cat-calls and abuse and hisses',
`clamoribus et conviciis et sibilis consectantur'.
27
`Graccho dicente silentium fuit: cum Claudio obstreperetur, audientiam facere
praeconem iussit.' Also Scipio Aemilianus was shouted down in a contio, Plut. Mor.
201F.
28
40C: `Qua re territi Clodiani silentio verba testium per biduum audiri passi sunt'.
29
Cic. Post red. sen. 26: ùt nemo cuiusquam conducti aut perditi vocem acerbam atque
inimicam bonis posset audire'.
48 The contio
138, also invokes the atmosphere of a Roman contio, where an irre-
verent crowd could prevent an unpopular speaker from being heard.
30
The vulnerability of the orator also explains the stress on the silence of
an audience as a sign of respect and approval. Thus, Cicero claimed that
in the contio which preceded the vote to recall him from the exile: `The
silence and the approval of all present were so intense that it seemed as if
nothing so popular had ever reached the ears of the Roman people',
Sest. 107, and continued: `What a silence there was, to hear the rest of
the leading men of the state when they spoke of me!', 108.
31
Similarly in
Mil. 91 Cicero stresses that in the contio following Clodius' death: `the
tribune M. Caelius had been given a silent hearing', before being
violently interrupted by political opponents.
32
Because of the absence of a disciplined political culture, or any
effective reinforcement of public order, it was often impossible to
conduct debates in public.
33
At the most extreme a speaker could be
physically assaulted by a hostile crowd, as often happened in the late
republic. Among the numerous instances are the brawl in 91, when
supporters of Livius Drusus injured the consul Philippus at a pre-
comitial contio.
34
Likewise, when the consul Piso challenged a contio
called by Cornelius in 67 against the senate, his fasces were broken and
stones were thrown at him, Asc. 58C.
Evidence such as this leaves us with the question of whether a
genuinely `democratic' institution could ever have functioned in the
polarised political climate of the late republic. Since it would have been
almost impossible to address a hostile crowd, let alone win them over,
we will have to redene the purpose and character of the contio. Having
lost its potential as a `democratic' forum, it appears to have been
reduced to its most basic function, a public stage for ofce-holding
politicians. That did not not diminish its importance. On the contrary,
in a society without any other mass media than word of mouth, public
meetings performed a vital function in disseminating political news and
30
`Tacete, quaeso, Quirites, plus ego quam vos quid rei publicae expediat intellego', Val.
Max. 3.7.33.
31
107: `. . . tanto silentio, tanta adprobatione omnium, nihil ut umquam videretur tam
populare ad populi Romani aures accidisse'; 108: `Quo silentio sunt auditi de me ceteri
principes civitatis.'
32
`vidistis . . . contionem gladiis disturbari, cum audiretur silentio M. Caelius, tribunus
plebis'. Cf. Cic. Cluent. 93.
33
Even in the senate debates were often so unruly that the speakers had difculty making
themselves heard, Cic. Q. Fr. 2.1.2; 3.2.2. The shouting might not be much different
from that in the contiones, cf. Cic. Q. Fr. 2.5.1: `clamore . . . prope contionali', and also
in the senate complete silence was considered a sign of approval and respect, Q. Fr.
2.1.1, describing how Rutilius Lupus was listened to `magno silentio.'
34
Florus 2.5.8; De vir. ill. 66.9; Val. Max. 9.5.2.
The contio 49
propaganda. In the late republic we nd contiones used for a variety of
purposes: apart from presenting new proposals, magistrates could attack
opponents or defend themselves against allegations. Cicero in Agr. 3.3
referred to accusations, which he wanted to counter in public. Contiones
could also be used to spread rumours and launch slogans.
35
Thus in 52
T. Munatius Plancus took the opportunity to spread a story about Milo
having abducted and imprisoned witnesses to the killing of Clodius,
Asc. 37C. A contio could also be used to put pressure on opponents and
faltering allies.
The contio obviously had a particular signicance and value for
`popular' leaders. It offered them an opportunity to demonstrate the
popular support on which they based their policies and claim to
inuence. As such they were effective symbolic manifestations of the
sovereignty of the people over the senate; as illustrated for example by
the gestures of Licinius Crassus and C. Gracchus, who abandoned the
Comitium in 145 and 122 in favour of the Forum. Likewise, in 58 the
consul Gabinius responded to the opposition of the senate by calling a
contio, where he gave vent to his anger in a furious speech, Cic. Sest. 28.
Contiones could also function as rallying-points for popular leaders,
strengthening support by rousing speeches and mass gatherings. They
might even be used to muster crowds in preparation for future action.
Thus, Munatius urged the people at the contio to turn up in large
numbers the following day when Milo was to be tried, ùt postero die
frequens adesset', Asc. 40C, cf. 52C. Such exhortations would probably
reach a wider circle than the audiences actually present; the news
launched on such occasions may have spread rapidly throughout the
city.
The contio became a powerful political weapon in the late republic,
used to attract favourable attention, put pressure on other politicians,
and generally score political points against opponents. But this use also
involved certain risks. Unpredictable as they were, contiones could back-
re and damage your prospects and reputation. If the organiser failed
to get a positive response, the whole purpose of the meeting would be
defeated, and instead of demonstrating his favor populi it would provide
his opponents with political ammunition. Thus in 59, when Caesar
held a less successful contio against his colleague Bibulus, Cicero
gleefully writes to Atticus that `. . . Caesar thought he might stir up a
contio with a speech into attacking Bibulus. After a long inammatory
harangue he could not raise a murmur', Att. 2.21.5.
36
Likewise, in Att.
35
Pina Polo (1996) 94113.
36
`. . . putarat Caesar oratione sua posse impelli contionem ut iret ad Bibulum. multa
seditiosissime diceret, vocem exprimere non potuit'.
50 The contio
4.2.3 Cicero gives a scathing account of the contio where even the
ìnmi' in a crowd addressed by Clodius failed to take him seriously
when he claimed that he had the pontiffs' support on the issue of
Cicero's house; they `were astonished, while others laughed at his
folly'.
37
The nature and purpose of contiones meant that certainly by the late
republic they were useful to a politician only if he could command a
supportive audience. Since there was no identity between audience and
voters, the whole point of calling a contio was to address a crowd which
could propagate one's ideas and boost one's public standing. Thus, the
internal logic of the institution was to develop into a stage-managed
political demonstration. In the polarised climate of the late republic an
effective contio required preparation, and it seems that politicians in-
creasingly relied on sympathetic crowds mobilised in advance which
also explains the partisan crowds that become such a prominent feature
in the last generation of the republic. In this period the crowds begin to
take on distinctly different political colouration according to who had
organised the meeting. The breakdown of the old political consensus
naturally led to new patterns of political participation, and given the
likely social bias of the traditional political crowd in the Forum, it is
difcult to see how `popular' leaders could conduct a successful contio
without mobilising their supporters among the masses. Therefore, as the
republic descended into anarchy, the contio naturally turned into a
partisan gathering of loyal supporters.
Apart from showing the absence of any real discussion, the passage
from Pro Cluentio, quoted above, also suggests the crowd's blank accep-
tance of the proposal put before it. Cicero links the apparent unanimity
of the contio to the suppression of open debate, but the presence of a
supportive crowd may have been a common feature of late republican
contiones. There seems to have been a general expectation that the
organiser of a contio would enjoy a sympathetic audience. In Sest. 105
Cicero contrasts the early popular leaders with Clodius; whereas the
former could mobilise a crowd by means of policies attractive to the
people, `largitio' and `spes commodi', Clodius, Cicero argues, had to
pay supporters to ll his meetings. Apparently, the presence of a
sympathetic crowd is taken for granted; only the means of achieving it is
discussed. And Cicero's comment on the crowd of ìnmi' at a contio
called by Clodius' brother Appius, Att. 4.2.3, which failed to accept
Clodius' claim to pontical backing, also gains its particular sting from
37
`Hic cum etiam illi inmi partim admirarentur partim irriderent hominis
amentiam . . .'
The contio 51
the assumption that the audience at one's own contiones was there to
provide support, not question your propositions.
It is therefore not surprising to nd very few examples of failed
contiones in the ancient record, despite the partisan nature of the sources
and their delight in the misfortune of opponents. For that reason we
should perhaps also be wary of stories suggesting contiones turning
against the magistrate in charge, or radically shifting allegiances during a
meeting; in some cases the interpretation leaves serious doubts about
the identity of both organiser and crowd.
Thus Granius Licinianus, 36.33, discusses the tribunes' attempt in 78
to restore the powers of their ofce, which had been diminished by
Sulla, stating that: `When the popular tribunes suggested that the
consuls should restore the tribunician powers, Lepidus was the rst to
refuse, and in a contio the majority agreed with his statement that there
was no need to restore the powers of the tribunes.'
38
It is not clear from
our text who had called the contio. We cannot automatically assume that
the tribunes had organised the event. It would seem unlikely that a
crowd assembled by the tribunes to put pressure on the consul could
have sided against them on this crucial issue.
39
More likely the consul,
in an attempt to counter the tribunes' motion, organised a contio to
show that the populus did not endorse the claims made in its name.
40
The impression of volatile crowds is very often the result of the
ancient chroniclers' schematic descriptions of collective behaviour,
referring to the `people' or the `crowd' without any qualications. An
example is Dio 39.289, which describes the dramatic events in 56
when the senate tried to prevent the consular candidatures of Pompey
and Crassus. The consul Lentulus Marcellinus had rst `brought the
multitude, which had thereupon rushed together, to a state of extreme
sorrow; no one had a word to say against him'. After the senators had
returned to the Curia, Clodius attacked Marcellinus before the people
38
`Verum ubi convenerant tribuni plebis, consules uti tribuniciam potestatem resti-
tuerent, negavit prior Lepidus, et in contione magna pars adsensa est dicenti non esse
utile restitui tribuniciam potestatem.'
39
There seems to have been considerable pressure for a restoration of the tribunician
powers. Thus, Cic. Cluent. 110, Leg. 3.26; Verr. 1.45, cf. Meier (1980) 140; Millar
(1998) 4971. Pompey gained great popularity when he implemented the reform, cf.
Gruen (1974) 25 n.57, and in 75 the tribunes' right to hold higher ofces had been
restored `magno populi studio', Asc. 67C.
40
Likewise, Sall. Hist. 3.48.8, on the tribune Sicinius from 76: `primus de potestate
tribunicia ausus, mussantibus vobis circumventus est . . .', does not, as recently
claimed by Pina Polo (1996) 142, refer to a meeting where the crowd failed to support
his endeavours on their behalf; it simply notes a general indifference among the people
when Sicinius was unjustly prosecuted. Cf. McGoshin (1994) 91, 28, translating the
passage: ànd even though L. Sicinius, the rst to raise the question of the tribunician
power, was circumvented while you only muttered about it'.
52 The contio
and tried unsuccessfully to get access to the senate house. Sur-
rounded by knights, Clodius called for help, and `many ran to the scene
bringing re and threatening to burn his oppressors'. The scenario is a
most unlikely one; a crowd moved within minutes from complete
agreement with the senate to violent support for Clodius. We are
probably dealing with competing groups of followers.
41
Plausibly, the
improvised meeting was rst dominated by the boni and knights who
were already present in the Forum; only later did Clodius' loyal
supporters arrive to rescue their leader.
In general the character of a contio appears to have been closer to a
partisan political manifestation than to a public debate. This is hardly
surprising since, as we saw, contiones did not offer the opportunity for
genuine discussion; as a rule politicians seem to have stayed away from
meetings held by their opponents.
42
Thus, we have very little evidence
for Roman politicians voluntarily attending or addressing hostile con-
tiones. By turning up they merely risked abuse or even physical assault
without having any real chance of making political gains. Thus, when
C. Cato in 59 ascended the Rostra to attack Pompey in a hostile contio
Cicero says that it was a wonder that he got away alive, Q. Fr. 1.2.15.
Another exception was Clodius' appearance at the contio of the boni,
organised by the senate in 57, Cic. Sest. 108. But that was clearly
intended as a gesture of deance, demonstrating his intransigence at the
senate's attempt to restore Cicero by calling the comitia centuriata.
Generally violent confrontations were a rare occurrence in contiones;
there was no point in provoking unnecessary disturbances in what were
essentially `party'-meetings.
The pre-comitial contiones, on the other hand, had a somewhat
different character. They were lled with voters about to attend the
ensuing comitia tributa or concilium plebis and might be the scenes of
open disagreements and confrontations. Livy, 34.1.4, gives a classic
description of a heated exchange of arguments in front of the assembled
people about to vote on the repeal of the Lex Oppia. Later the pre-
comitial contiones turned into virtual battlegrounds; in post-Gracchan
times the course of many meetings shows the impossibility of conducting
proper debates in public. Thus, in 67 the senators railed against the Lex
Gabinia to little avail since they could not even get a hearing, Plut.
Pomp. 25.36.
41
Val. Max. 6.2.6 simplistically states that Marcellinus addressed ùniversus populus'.
42
Thus, when Cicero turned up for the contio held by Flavius in 60, it was to lend
qualied support for the proposed land bill, to which he suggested some
modications, Att. 1.19.4. Cicero could therefore claim the good will of the crowd
`secunda contionis voluntate', which as a matter of course had backed Flavius'
scheme.
The contio 53
Cicero claims that riots and violence often broke out in contiones, Sest.
77, but that statement may be too general. A closer look at the evidence
suggests that most clashes took place at the pre-comitial contiones, which
were attended by political opponents and their respective followers.
Thus, the contio before Minucius Rufus' attempt to have the Lex Rubria
repealed in 121 ended in a clash with C. Gracchus, who had arrived
with a large crowd. Likewise in 91, when Philippus opposed Drusus'
laws and was injured by one of his supporters. Another example, also
showing the absence of a plebs contionalis, is provided by Metellus'
attempt to have Pompey entrusted with special powers to crush the
Catalinarian conspiracy. Plutarch, Cato Min. 269, describes the as-
sembly, which descended into an open brawl between the supporters
brought along by Metellus and those raised by his opponent Cato on
behalf of the senate. When Metellus' men had been routed, the `people'
suddenly appear in the story, strongly supporting Cato. But this is
obviously a reference to Cato's victorious followers, as indicated by their
denition as `the better citizens' in 27.6. Therefore, in this meeting we
nd no trace of a `political crowd', only the partisans mobilised by
various political leaders.
Contiones called ad hoc generally seem to have been one-sided,
organised events masquerading as the assembled Roman people. The
`staged' character of the meetings offered magistrates an opportunity to
pressurise and embarrass political opponents in public. An invitation to
a hostile contio was a `no-win' situation for the politician in question.
Ignoring a challenge from a magistrate could be used against oneself,
making one liable to accusations of personal cowardice and disrespect
for the sovereign people. Thus, Cicero taunted Rullus for not turning
up at his contio, Agr. 3.1. Attending, however, meant that one had to
face a hostile crowd organised by an opponent who could thus attack
from a much stronger position, supported by their shouts and jeers.
In 121 Cn. Carbo called Scipio Aemilianus before a contio to inter-
rogate him about his stand on the killing of Ti. Gracchus. Defending the
act, Scipio was abused by the audience and in return accused it of being
the stepchildren of Italy.
43
Carbo, it seems, had organised a sympathetic
crowd to put pressure on Scipio and underpin his claim of popular
outrage. Scipio Nasica and other senators were put in a similar situation,
when they were called before a hostile contio after Gracchus' death,
Diod. 34/35.33.7.
In a letter to Atticus, 1.14.1, Cicero describes Pompey's
43
Cic. Mil. 8; Val. Max. 6.2.3; Vell. 2.4.4; De vir. ill. 58.8; Plut. Mor. 201F; Polyaenus
Strategemata 8.16.5. According to Plut. Ti. Grac. 21.5, Flaccus and C. Gracchus
interrogated Scipio Aemilianus in front of the people.
54 The contio
unsuccessful appearance at a contio held on the issue of Clodius'
prosecution after the Bona Dea scandal in 61. He had been summoned
by the tribune Fuus who had mobilised a large crowd in the Circus
Flaminius, including people from the market. Here Pompey was
quizzed over the selection of jurors for Clodius' trial. In 56 the consul
Marcellinus called Pompey and Crassus before an optimate contio,
pressing them to reveal whether they intended to run for the consulship,
Plut. Pomp. 51.56. Similar interrogations in front of hostile audiences
took place before the exiling of Cicero, when: `Clodius brought them
[Hortensius and Curio] before the people, where they were soundly
belaboured for their mission by some appointed agents', Dio 38.16.5.
44
In this light it is hardly surprising that Rullus declined the invitation
to appear at Cicero's contio and defend his land reform, preferring
instead to challenge the consul to a debate in a contio of his own, Plut.
Cic. 12.5. The manoeuvrings around this bill make little sense if
contiones in general were held in front of a regular plebs contionalis: they
suggest a particular `homeground' advantage to those contiones you had
called yourself.
45
The rules dealing with `competing' contiones point in
the same direction. Gellius, 13.16, tells us that in cases where several
contiones had been called simultaneously, the rst to be announced
would take precedence. Again the rule underlines the political impor-
tance of calling one's own meetings, rather than appearing at others'.
The staged character of the contio throws doubt on the role of oratory
as a political weapon. In itself the human voice is hardly an effective
medium of mass communication; and as noted above, it may even have
been practically impossible to address a crowd which was not already well
disposed towards the speaker. Contiones were clearly important in terms
of spreading news and slogans; but again this function also relied on a
favourable crowd. Successful use of rhetoric as a means of swaying voters
in one direction or another may therefore have been a rare occurrence.
Cicero's obstruction of the rogatio Servilia is interesting in this
context. It is the only instance from the late republic where oratory
applied in a contional address would appear to have prevented the
passing of a `popular' bill.
46
In 63 the tribune Servilius Rullus put
forward a carefully prepared plan for the distribution of land to the
44
Cf. Dio 38.16.6; Cic. Dom. 40; Har. resp. 48. In 59 Caesar used a similar strategy when
he put Bibulus before a hostile contio, Dio 38.4.23, obviously involving a different
audience from the one that had attended the contio recorded in Cic. Att. 2.21.5.
45
Nor do they t into a model of issue-driven participation in meetings; presumably those
with a particular interest in Rullus' bill would have turned up to contiones held both for
and against the reform, in which case the audiences would have been identical.
46
E.g. Gruen (1974) 395, who suggested that the freeborn plebeians `may well have been
led astray by the obfuscating rhetoric of M. Cicero'.
The contio 55
poor. Cicero opposed it vigorously, delivering four speeches, two in the
senate and two at contiones. He later claimed to have obstructed the bill
by the powers of his oratory, turning popular opinion against the bill; a
view which reappears in Plutarch, Cic. 12.5.
47
However, the bill was not
defeated in the comitia, as Pliny claims, NH 7.117, since it never reached
the assembly. The tribune Caecilius Rufus had threatened a veto, thus
forcing Rullus to withdraw the proposal, Cic. Sull. 65.
48
Cicero's public
speeches may therefore have had no direct inuence on the failure of the
bill.
Cicero's speeches against the rogatio Servilia obviously received a
favourable response. But we should bear in mind that he was addressing
contiones convened at his own initiative; presumably the audience was
already on his side when the meeting began. We have no reason to
believe that it represented a broad section of the plebs, including
potential future beneciaries of Rullus' bill. The meeting at which
Cicero gave his longest speech was in fact the traditional opening contio,
in which the new consul thanked the people for his election, cf. Cic. Fin.
2.74. Thus in Agr. 2.14 Cicero expresses his gratitude to the crowd for
their support in the election. Since Cicero had been elected by the
voters of the rst two classes alone, the implication would be that the
audience consisted of supportive boni, who had turned up to celebrate
the new consul at this rst public address. Evidently its composition was
quite different from the supportive crowd to whom Rullus had rst
presented his bill; in 2.13 Cicero explicitly states that no one in that
audience had understood what the law was really about.
The purpose of Cicero's speeches must be reconsidered. On one level
they were a public demonstration against Rullus and his backers. They
were also an attempt to manipulate public opinion in those quarters
which mattered to Cicero, that is the boni or propertied classes, alerting
them to the ominous threats of Rullus' seemingly moderate bill. Finally,
we cannot exclude the possibility that the real target of Cicero's series of
addresses may have been the tribunes, who had the power to block the
proposal. They sat in the senate and would also have been susceptible to
pressure from the boni, whom Cicero was trying to inuence.
It seems that the practical role of oratory may have to be redened.
Public oratory would in the nature of things very often have been
preaching to the converted. Its effect lay not in swaying the minds of the
numerically quite insignicant audiences who could be reached in this
way; its importance may have lain in the rousing effect it had on the
faithful, who were encouraged to continue campaigning on one's
47
Cic. Rab. perd. 32; Pis. 4; Fam. 13.4.2.
48
Stockton (1971) 91; Pani (1997) 21617.
56 The contio
behalf.
49
Indirectly public speeches may also have inuenced broader
sections of the population by providing soundbites and slogans which
might become more widely known.
50
The general tendency of the organisers to have the crowds behind
them, whatever their political stance or afliations, may explain why
different meetings could take such radically contrasting positions on the
same issues. The changing character of contiones according to which
politicians were responsible for organising them is illustrated by the
meetings concerned with Cicero's restoration from exile. In his speech
for Sestius, Cicero sharply contrasts the contiones organised by Clodius
and those held by his optimate supporters before his recall from exile:
That villainous gladiator held many meetings about me to which nobody came
unless bought and corrupt. No honest man could stand the sight of his ugly face
or the sound of his madman's voice. Those meetings of blackguards had to be
disorderly. Publius Lentulus as Consul held a meeting, likewise about me. The
Roman People attended in force. All classes, the whole of Italy, took part in that
meeting. He put the case very impressively and eloquently; and from the silence
and universal approval it appeared that nothing so popular had ever fallen upon
the ears of the Roman People.
51
Cicero uses the fact that the participants were different at meetings
organised by different political groupings to draw a distinction between
real and distorted contiones. He condemns the contiones of his opponents
as unrepresentative and attended merely by a hired rabble.
52
Those held
in his favour, on the other hand, reected the true feelings of the people,
cf. e.g. Mil. 3: `reliqua vero multitudine, quae quidem est civium, tota
nostra est'. Likewise in Sest. 108, Cicero mentions the `verum
populum'. And in Mil. 91 the audience of an optimate contio is
logically referred to as the `populus Romanus'.
53
49
On a different level contiones held an important symbolic position as the ofcial
medium of communication between leaders of the state and the sovereign Roman
people. As we have seen, the crowds by denition represented the entire populus
Romanus, and any positive reaction received here therefore had considerable propa-
ganda value to the politicians concerned.
50
Speeches delivered at contiones might also be published and circulate outside Rome
among the municipales, cf. Millar (1998) 29, 126, 145, 195.
51
Shackleton Bailey (1991) 1867, Sest. 1067: `Habitae sunt multae de me a gladiatore
sceleratissimo, ad quas nemo adibat incorruptus, nemo integer; nemo illum foedum
vultum aspicere, nemo furialem vocem bonus audire poterat. Erant illae contiones
perditorum hominum necessario turbulentae. Habuit de eodem me P. Lentulus consul
contionem; concursus est populi Romani factus, omnes ordines, tota illa contione Italia
constitit. Egit causam summa cum gravitate copiaque dicendi tanto silentio, tanta
adprobatione omnium, nihil ut umquam videretur tam populare ad populi Romani
aures accidisse.'
52
Cf. Att. 4.3.4: `contiones turbulentae Metelli, temerariae Appi, furosissimae Publi', and
Sest. 104.
53
Cf. e.g. Cic. Fam. 1.4.2; Q. Fr. 2.4.6.
The contio 57
There were obvious differences in the social composition of the
audiences. Thus, in Dom. 545 Cicero describes a `conventu virorum
bonorum', which was attacked by Clodius' people, and at Lentulus'
meeting a large contingent had been called up in the municipalities. But
there was probably no essential difference between what Cicero saw as
the `people's contiones' and the meetings of his enemies. The magistrate
who organised them had decided the theme, selected the speakers and,
we may assume, made every effort to secure a supportive crowd in
advance. Therefore, when Cicero revels in the positive response he or
his cause had met in a contio, it was all part of the same transparent
ction of real and false contiones. Those held in his favour may have been
no less planned and organised than those campaigning against him. In
contiones the audience invariably applauded; that was the function of a
public meeting, whatever its political colouring.
Since contiones increasingly became organised political events, the
turnout could become a measure of success and the level of attendance
a source of personal pride. Thus, Cicero boasts that `contionem . . .
maximam' had been held in his support, Phil. 14.16. Likewise in Phil.
4.1; 6.18; 7.22, the magnitude of the audience he was addressing is
stressed with obvious delight. And in the published version of his second
speech against Rullus, 2.103, Cicero carefully notes the large crowd he
was able to command and how well it boded for his consulship.
54
On the
other hand, small attendance at a contio could be used against an
opponent. Thus, Cicero mocks the tiny audiences gathered by Clodius
in his speech ìn Clodium et Curionem'.
55
These points, making little
sense in the context of a permanent plebs contionalis, suggest that the
turnout was not constant, but could be a sign of prestige. Turning up for
someone's contio was probably not seen as a show of `political interest'
but of afliation and support.
This line of argument takes us to the issue of mobilisation. How
were crowds organised for contiones and who was called upon? First,
however, we will have to recognise the serious limitations to our
knowledge on this aspect of late republican politics. The sources leave
us in almost complete darkness as to the composition of, for example,
the crowds addressed by Ti. Gracchus in 133 or the audience backing
the consul in 78. Likewise, we have no idea how the crowd had gathered
which greeted Cicero so warmly in 63 at his rst contio after taking
ofce.
54
`. . . Quirites, ut, qualis vos hodierno die maxima contione mihi pro salute vestra
praebuistis . . .'
55
Àccesserunt ita pauci, ut eum non ad contionem, sed sponsum diceres advocasse',
Schol. Bob. 88 (St) frg. 15.
58 The contio
Not least against this evidential background it seems paramount that
we avoid a model of popular mobilisation, which assumes a single
unied pattern. There were probably a variety of methods and strate-
gies open to Roman politicians, ranging from the use of highly orga-
nised groups to a more improvised encouragement of people in a
neighbourhood or social circle. A rare example of a less well-organised
crowd has already been mentioned above; in 59 Caesar's contio against
Bibulus did not go according to plan. Another example comes from
Brutus 305, where Cicero says that in 90 the tribune C. Curio was
deserted by the crowd in the Forum during a contio he had convened.
These instances suggest that not all crowds may have been tightly
disciplined bands of personal followers, while at the other end of the
scale we nd Clodius' mobilisation of crowds, which emerges from our
deeply hostile sources as an extreme and unprecedented case of
popular organisation.
Cicero describes in detail how Clodius drafted supporters by the
Tribunal Aurelium in the Forum Romanum: Ànd with the same
consuls looking on, a levy of slaves was held in front of the Tribunal of
Aurelius on the pretext of forming collegia; men were enlisted street by
street, formed into squads, and incited to force, acts of violence,
murder and robbery', Sest. 34.
56
This as well as other references to
collegia and neighbourhoods suggests the existence of an effective
network for mobilisation based on the associations of the plebs; an issue
to which we shall return in the discussion of the comitia. Clodius was
assisted by intermediate leaders, duces, who maintained direct contact
with the plebs, e.g. Cic. Dom. 12; 89. A number of these assistants are
named in our sources.
57
The existence of such professional organisers
is already attested at the trial of Cornelius in 65. His accusers were
surrounded by `notis operarum ducibus', Asc. 59C, and Manilius
disturbed the trial `per operarum duces', Asc. 60C. Similarly in connec-
tion with the Catiline conspiracy Sallust, Cat. 50.1, mentions: `leaders
of crowds who were wont to cause public disturbances for hire', and
Cicero, Cat. 4.17, refers to rumours that a `pimp of Lentulus is making
the rounds of the shops, hoping to buy the support of the poor and
ignorant'.
58
56
Ìsdemque consulibus inspectantibus servorum dilectus habebatur pro tribunali Aurelio
nomine collegiorum, cum vicatim homines conscriberentur, decuriarentur, ad vim, ad
manus, ad caedem, ad direptionem incitarentur.' Cf. Cic. Dom. 54; Post red. Quir. 13.
57
Favory (1976); Flambard (1977) 12631; Benner (1987) 15669.
58
`duces multitudinum qui pretio rem publicam vexare soliti sunt', and àuditum est
lenonem quendam Lentuli concursare circum tabernas pretio sperare sollicitari posse
animos egentium atque imperitorum.' Note also the reference in Comm. pet. 51 to
inuential plebeians `qui contiones tenent'.
The contio 59
Cicero employs a whole arsenal of insults against Clodius' contiones.
The standard terms of abuse are ìnmi', `perditi', ègentes', and
`facinerosi', suggesting a disreputable crowd of the poor, the depraved
and the criminal.
59
Whatever the reality of these allegations, it is evident
that those attending Clodius' meetings were not men of property but
visibly belonged to a lower probably working-class stratum of society.
Tabernarii and artisans have often been identied as the core suppor-
ters of Clodius and other `popular' leaders.
60
Considering the fact that
they made up a very large part of Rome's working population, this
suggestion is hardly controversial. Explicit references are, however,
remarkably rare, which may be due to their self-evident character.
Catiline is reported to have mobilised artisans, App. B. Civ. 2.17, cf.
Cic. Cat. 4.17, quoted above. Likewise, Sergius, an ally of Clodius, is
labelled a `concitator tabernariorum' by Cicero, Dom. 13. But there are
also indications that raising a large crowd of shop-keepers was not that
easy after all. Thus, the repeated iustitium and closure of the tabernae,
issued by Clodius, was a radical measure which suggests that it took
more than a little persuasion to make tabernarii leave their shops and
take part in political events, no matter how important.
61
The role of freedmen in `popular' meetings has also, as we have seen,
been emphasised by modern historians. Again this is a fairly safe
assumption, since their share of the capital's population would have
been quite considerable. And as was the case with the tabernarii explicit
references to this category are not very common; perhaps their partici-
pation was also too obvious to attract much attention. Occasionally we
nd indirect suggestions of freedmen in the crowds. Thus, foreign
origins and libertine status were implied in Scipio Aemilianus' accusa-
tion that the crowd were the stepchildren of Italy (see above p. 53). And
Cicero, Flacc. 17, claimed that foreigners, that is freedmen of Greek
extraction, often disturbed public meetings in Rome. But that was not
part of the standard repertoire of insults against hostile contiones.
Most references to the legal status of members of the audience
concern slaves. Thus, Cicero, Sest. 34, claims that Clodius conducted
59
Cic. Mil. 36: `servorum et egentium civium et facinerosorum armis'; 95: `plebem et
inmam multitudinem'; Dom. 89: `multitudinem hominum ex servis, ex conductis, ex
facinerosis, ex egentibus congregatam'; 96: `perditi'; 45: `conductos, sicarios, egentes,
perditos'; Post red. sen. 26: `conducti et perditi'; Sest. 95: `carcerem totum in forum
effudit'; 23: `sicarii'; 76: `latrones'; Vat. 21: `vi perditorum hominum incitata'; 40:
`Clodianas operas et facinerosorum hominum et perditorum manum'; Pis. 26: `vis
latrocinii' and 11; 30; 64.
60
Treggiari (1969) 175.
61
Cic. Dom.; 54; 8990; Acad. 2.144; Asc. 401C, 52C.
60 The contio
`servorum dilectus'.
62
We have no way of telling whether slaves really
took part in the meetings. Some scholars believe it to be derogatory
references to freedmen, but on several instances Cicero is quite unequiv-
ocal, e.g. Mil. 76 and Dom. 54, where he claims that Clodius mobilised
`not only free men but also slaves'.
63
On principle there is no cogent
reason why slaves should have been completely absent from contiones;
there was no identity check to exclude non-citizens from taking part.
Small wonder, therefore, if the organiser or some of his followers should
bring a few slaves along to boost attendance levels. The practice
obviously invited censure and could be used to discredit opponents,
thus explaining why slaves feature in our sources so much more often
than freedmen.
The other main objection to the contiones of Clodius focused on the
claimed involvement of money in his mobilisation of crowds. It recurs
frequently in Cicero, who constantly refers to the hired operae of
Clodius.
64
Modern observers have often followed Cicero's condemna-
tion and seen the practice as an indication of the degree of political
disintegration reached in the late republic. But by dismissing out of
hand this practice simply as corruption we miss some important impli-
cations of Clodius' remuneration for political services. Modern views of
the democratic process as ideally untainted by any involvement of
money are misplaced in a Roman context. The absence of Greek-style
remuneration provided by the state was crucial to the ability of the
Roman people to take part in day-to-day politics. The implication was
that a broader social representation in the assembly was possible only if
economic compensation to the lower classes was provided by other
sources. Or in other words, to achieve a regular attendance by members
of the lower plebs the politicians would have to make the necessary
outlay themselves.
65
Clodius' great innovation thus lay in his mobilisation of a permanent
crowd, which could be relied upon to turn up frequently and at short
notice. This was of crucial importance if a popularis was to maintain a
constant political presence and visibility in Rome. Contiones were only
62
Cic. Acad. 2.144; Sest. 95; Post. red. sen. 33; Dom. 5; 89; Mil. 36, 37; Pis. 11; 23; 30; 57;
Att. 1.16.5; 4.3.4, cf. Asc. 32C.
63
`non modo liberos sed etiam servos'. Cf. Dom. 79: `concilio advocato, conductis operis
non solum egentium, sed etiam servorum'. Treggiari (1969) 1724, 2656.
64
Cic. Sest. 38; 106; 127; Dom. 45; 79; 89. Cf. Brunt (1966) 235.
65
Brunt (1988) 434 has expressed doubts about Clodius' nancial ability to hire crowds.
Firstly, however, the crowds in question were not very large; secondly the pay rate was
probably quite basic; and thirdly Clodius' personal fortune would have been one of the
largest in Rome, also enabling him to expand his Palatine mansion several times.
Treggiari (1969) 175 suggests that Crassus may have sponsored Clodius' crowds, cf.
Cic. Q. Fr. 2.3.4.
The contio 61
one element in this type of campaigning, others were informal demon-
strations and attacks on opponents, but they were important symboli-
cally as the people's ofcial stage and more practically as rallying points
for loyal supporters.
The crowds of his senatorial opponents are more difcult to assess,
hidden as they often are under the guise of the `real people'. It is possible
that the senate, on those occasions where it felt a need to manifest its
position in public, relied simply on the people who were around in the
Forum to provide a sympathetic audience. As we saw, the crowds
mobilised in this way would not have been the small traders, identied
by Mommsen and Meier. More likely they were equites and boni, like the
respectable crowd which rushed from the Forum to the support of
Marcellinus.
Despite the availability of sympathetic crowds in the Forum, the late
republic saw a gradual change to the pro-senatorial meetings. In
response to the still more efcient mobilisation conducted by the
`populares', crowds were increasingly organised in advance to support
the senate's cause also. The optimate contiones may therefore have been
less different from those of Clodius than Cicero is willing to admit.
Leading politicians of every colour now had personal guards and a
following ready to be called into action. Milo's gang of strong-arm men
is well known, but may be part of a much more general mobilisation. As
we saw, Cato mobilised men for the comitia, Plut. Cato Min. 268. And
an indication that these senatorial bands were used also in contiones is
found in Cic. Att. 2.16.1, which mentions that the 5 per cent tax could
probably be `swept away by the shouts of our footmen at a single scratch
assembly'.
66
Likewise, Cicero refers to the use of `nostris' in a clash with
Clodius' operae at the trial of Milo in 56, Cic. Q. Fr. 2.3.2, and in Sest.
27 he mentions `my own gang', òperae meae'.
Private guards are well documented for many late republican politi-
cians. Their composition may have varied, but nothing suggests they
were temporary measures formed ad hoc. Cicero's guard can be traced
over a period of more than ve years. It was made up of knights and
Reatines,
67
and according to Plut. Cic. 16, it was so large that à great
part of the Forum was occupied when he entered with his escort'.
Gladiators could also be enrolled in the guards of the nobles.
68
Likewise, Clodius' private sponsorship of contiones may not have been
as exceptional as Cicero would like us to believe; other politicians of
66
`contiuncula clamore pedisequorum nostrorum'.
67
On the social background of Cicero's attendants, Cic. Phil. 2.16; Cat. 3.5; Sall. Cat.
26.2.
68
Lintott (1968) 835.
62 The contio
different allegiances probably used this or similar strategies to some
extent. Milo brought bribed people to Rome, App. B. Civ. 2.79, and an
element of payment may also have been present in the maintenance of
the private guards. Thus, Crassus' famous saying that: `no one is rich
enough who cannot feed an army out of his income', may be a reference
to private guards rather than armies.
69
Household slaves were probably
often raised for this purpose, as happened in 133 when they were used
by Octavius, Plut. Ti. Grac. 18.2.
In conclusion, this discussion of the late republican contio would
suggest that the institution underwent important changes during the
period in question. Before the late second century the vast majority of
citizens in the city probably never appeared in the contiones, which may
traditionally have been gatherings of the boni, for whom participation in
politics was a natural pursuit and pastime. With the emergence of the
populares wider sections of the population would have been drawn into
the world of politics. The meetings in turn acquired a new function and
character. In the polarised political climate of the late second century a
new type of contio developed: the planned `party'-event. The result was a
more frequent use of popular meetings, which were increasingly lled by
pre-organised crowds. The `politicisation' of the contio may have had a
close parallel in contemporary developments of the legislative comitia,
which will be investigated in the following chapter.
69
Cic. Parad. 45, `. . . neminem satis esse divitem, nisi qui exercitum alere posset suis
fructibus', cf. Cic. Off. 1.25; Pliny NH 33.134; Plut. Crass. 2.7. See Nowak (1973) 97.
63
4 Legislative assemblies
When the ofcial debate, the suasio/dissuasio, drew to a close, the contio
was dissolved and the comitia tributa or the concilium plebis was called.
1
The assembled crowd was divided into tribus, and after a short prayer by
the presiding magistrate, the successive voting of the tribus could begin.
When a majority of tribus had accepted the proposal, it became law and
binding for the entire Roman people. In this process it is tempting to see
a direct democracy at work, as scholars have increasingly done in recent
years. But while the system was no doubt direct, the question remains of
how democratic the republican assemblies were in reality.
There were important formal limitations to the people's inuence.
Thus, in a `constitutional' perspective the passive, `reective' role of the
assembly is conspicuous. The comitia could only reject or approve
proposals put before it by a magistrate. It could take no initiatives of its
own nor suggest emendations or additions to bills. Moreover, the
principle of corporate voting, though in itself not undemocratic, offered
a perfect means of reconciling the principle of equal political rights with
the elite's de facto domination of the political process. Without violating
the formal political equality of Roman cives optimo iure this peculiar form
of voting enabled the elite to give different weight to individual votes.
Thus, the tribal assembly was strongly biased against the urban plebs,
which was inscribed in only four of the thirty-ve tribus, and against
freedmen who were also allocated to the urban tribus. As we shall see,
this disparity in inuence had a crucial impact on the workings of the
assembly.
But despite the inequality built into the institutional framework, the
comitia tributa and the concilium plebis did offer Roman citizens direct
access to the political process and a chance to have the last word in any
matter of legislation. The `direct' element in ancient democracy,
however, turns out to be crucial; in Rome it meant that only a very small
part of the citizenry could exercise their political rights. This chapter,
1
For the procedure see e.g. Cic. Flacc. 15; Asc. 71C.
therefore, will focus on the practical aspects of legislation, which are
viewed in the context of the quantitative estimates made above. The
specic issues to be discussed are (1) the patterns of voting in the
assembly, (2) the question of clientela and elite control, (3) the rise of
the populares and the emergence of popular mobilisation.
The degree of `democracy' in the legislative process must be assessed
against the actual voter behaviour observed in the assembly. Several
scholars have made the important observation that the comitia never
seems to reject any proposals put before it.
2
As Mommsen noted: Ìn
der Regel standen die Leute da und sagten ja zu allen Dingen'.
3
This
apparent co-operativeness has in turn been seen as an indication of an
impotent and tightly controlled assembly, which in effect was little more
than a rubber stamp lending symbolic legitimacy to decisions made by
the political class. As I shall argue below, other interpretations of this
voting pattern are possible. First, however, let us briey consider the
validity of this claim.
Rejections by the assembly are not entirely unknown, although they
remain extremely rare. Altogether ve certain cases are documented
from the last 150 years or so of the Roman republic.
4
In 167 a proposed
triumph for Aemilius Paullus was about to be turned down in the
comitia, when the consul intervened and persuaded the crowd to pass
the bill. The situation was exceptional. Paullus had made himself
unpopular with his soldiers, and one of the ofcers, the military tribune
Ser. Sulpicius Galba, organised the opposition to his triumph. The
mobilisation of the soldiers was so effective that they alone lled the
entire Area Capitolina.
5
In other cases the bill was never passed. In 149
an investigation into the conduct of the very same Galba in Spain had
been proposed by a tribune with the backing of the senate. The bill,
however, was defeated in the comitia after a passionate plea from Galba.
6
A few years later, in 145, C. Licinius Crassus proposed that appoint-
ments to the major priesthoods be transferred to the tribus. The bill was
rejected after a heated debate in front of the assembled voters.
7
In 130 a
proposal by C. Papirius Carbo to allow tribunician iteration met a
similar fate. The bill was faced with strong senatorial opposition and
2
Nippel (1981) 76; Burckhardt (1990); Eder (1991) 179; Flaig (1995) 801.
3
Mommsen (18545) I, 810.
4
Most famously in 200 BC the comitia centuriata refused to accept the war against
Macedonia, advocated by the senate. Only the second time round, after renewed
pressure had been applied, was the bill passed by the assembly.
5
Livy 45.35.836.6; Plut. Aem. Paull. 31.12.
6
Livy Per. 49; Cic. Brut. 80.
7
Cic. Lael. 96; Brut. 83.
Legislative assemblies 65
eventually turned down by the comitia.
8
Finally, in 104 an agrarian bill
put forward by L. Marcius Philippus was rejected in the assembly, Cic.
Off. 2.73.
After 104 no legislative defeat is securely documented. Some cases
have been interpreted in this way, but the evidence is weak. Thus, for
example, Cato's failed attempt in 61 to make equestrian jurors in
repetundae trials liable to prosecution for bribery. The bill may not have
been rejected in the assembly. Cicero, Att. 1.18.3, simply states that
`nulla lex perlata'; it probably succumbed to combined opposition from
the knights and parts of the senate.
9
Likewise, it has been suggested that
the failed citizenship bill of C. Gracchus was turned down by the
assembly.
10
But that overlooks the evidence of Appian, B. Civ. 1.101,
who tells us that Livius Drusus vetoed the bill, thus preventing it from
reaching a vote. Still, it cannot, of course, be ruled out that the alleged
unpopularity of the proposal played a part in its eventual failure.
Similarly, in 63 the rogatio Servilia probably never came before the
comitia; as we have seen, the reported threat of a tribunician veto would
imply that it was withdrawn before reaching a vote.
Since we know of a large number of proposals which failed to become
law, it is still possible to construct a scenario which allows the people a
certain input on legislation. In principle, the apparent tractability of the
assembly may in fact have been due to the nely tuned political instincts
of the magistrates who carefully considered the response of the people
before putting a proposal forward for the nal vote. Or, in other words,
proposals which proved unpopular at the presentation may simply have
been withdrawn or modied by their sponsors in order to avoid defeat
and humiliation. On this interpretation the people would maintain a
certain albeit indirect inuence on legislation.
11
The small scale of Roman politics represents a fundamental difculty
for this model. It means that a contio might not be representative of the
general views held by the electorate. Given the scale and diversity of the
urban masses, there was no homogeneous populus Romanus which could
reply with one voice to a new proposal. Moreover, since several weeks
would pass between the presentation of a bill and the nal vote, the
composition of the crowds attending on these occasions might be very
different. The reception of a bill in the rst contio would therefore not
8
Cic. Lael. 96; Livy Per. 59. Cf. Broughton (19501) I, 502.
9
Contra Gruen (1974) 241.
10
Thommen (1989) 77.
11
Thus, Flaig (1995) 936, followed by Pani (1997) 155, who claims that proposals were
changed according to their reception at the rst contio.
have given a reliable indication of its chances of eventually becoming
law.
There are, moreover, very few certain cases on record where the
public response forced the withdrawal of a bill between the rogatio and
the vote, and often this happened under the threat of a tribunician veto.
Again, the rogatio Servilia may serve as an illustration. The bill's failure
has generally been attributed to its unpopularity among the plebs; a
claim, however, which is based entirely on the testimony of Cicero
himself and of later writers inuenced by him. Cicero also reveals that
Caecilius Rufus had threatened a veto; the mood of the people may
therefore have been immaterial to its failure.
12
Indeed it does seem
unlikely that Rullus' bill could be obstructed simply by two speeches
delivered before a small crowd of Cicero's supporters. Another example
is Cornelius' proposed ban on loans to foreign delegates. The bill was
withdrawn, not because of its unpopularity but due to strong senatorial
opposition, Asc. 578C.
Many other bills known to have failed were probably never formally
proposed. They include, to mention but a few, the Gracchan measures
concerning military conscription, C. Gracchus' and Manilius' reforms
of the comitia centuriata, the proposed abolition of debt in 63, and
Lepidus' demand for a grain law in 78, Gran. Lic. 36.345. Often such
policy suggestions may have been mere kite-ying to test reactions in
various quarters. If they caused an outcry from inuential groups they
were likely to be quietly dropped. Even if the opposition might not
always have been insurmountable, it could still have been both contro-
versial and potentially hazardous to ignore it. Senatorial opposition
often played a key role. Thus, Laelius' agrarian reform from 140(?)
probably failed because of lack of support in his own political hinterland,
Plut. Ti. Grac. 8.4. `Popular' proposals also frequently succumbed to
senatorial pressure.
In some instances popular dislike may have played a part. That seems
to have been the case with the enfranchisement plans of Fulvius
Flaccus, C. Gracchus and the younger Drusus in 125, 122 and 91
respectively, of which only Gracchus' was formally proposed. But few if
any of the decisions to ditch a proposal can be ascribed solely to a lack
of popular support. Many factors contributed to bring a bill down;
popular opinion may not always have been the most important one. As
a rule the decisive pressure would have been applied by the senate or
other magistrates.
12
Cf. Pani (1997) 217. This crucial aspect has been overlooked by Bell (1997).
The comitia thus emerges from this brief survey as an acclamatory
body, which hardly ever exercised the formal choices put before it.
13
Its
stand on an issue may have had little impact on the fate of a bill or
even of informal proposals. The common strategy to explain this voting
behaviour has involved the clientela model, which operates with a web of
personal ties between the plebs and the senatorial elite. The strength and
extent of these ties effectively reduced the `democratic' element of the
political system to little more than a formality. Since individual voters
were not in a position to exert any independent choice, the passing of
bills became a matter of routine.
This analysis is of course purely structural; when a chronological
aspect is added to the model an entirely new picture appears. For with
the emergence of the populares in the second half of the second century
the political implications of the assembly's apparent co-operativeness
are turned upside down. Laws which challenged the vital interests of the
senate and the very foundations of its authority were now passed with
almost the same regularity as those approved by the senate. It seems
nothing if not paradoxical that while the nature of the proposed legisla-
tion was transformed, the assembly maintained its habit of passing
virtually every bill put before it. We are therefore witnessing in this
period a shift, almost overnight, from seemingly tight senatorial control
to what appears almost as permanent revolution.
This change has been explained by a supposed weakening of elite
patronage, in turn exploited by the `popular' politicians. Typical is
Bernstein's view that: Às the bonds of clientela loosened, the voters
of the concilium plebis exercised greater freedom, providing new scope
for the ambitions of Roman politicians . . .'
14
Likewise, Wallace-
Hadrill spoke of òld structures of deference' breaking down, while
Vanderbroeck claimed a `relaxation of old vertical ties among a large
group of the lower orders'.
15
However, the collapse of social ties on this
13
Interesting in this context also is the fragment of Fannius' speech against C. Gracchus'
citizenship bill, where the consul warned that: `Si Latinis civitatem dederitis, credo,
existimatis vos ita, ut nunc constitisse, in contione habituros locum aut ludis et festis
diebus interfuturos. Nonne illos omnia occupaturos putatis', ORF
3
frg. 3. Notably,
Fannius does not mention the comitia but the contio, which features in the context of
games and festivals, i.e. as one among various other diversions open to the upper
echelons of society. The contio thus appears from this passage less as a `political' forum
than as a form of public entertainment.
14
Bernstein, (1978) 193.
15
Wallace-Hadrill (1989) 66; Vanderbroeck (1987) 162, cf. 159. Similarly, Bleicken
(1975) 2789; Millar (1984) 17; Eder (1991) 192; Burckhardt (1988) 60; Linderski
(1985) 90; Nippel (1995) 78. According to Brunt (1988) 32, cliental ties had already
weakened in the third century.
interpretation becomes very abrupt indeed; that draws attention to the
foundations of this theory.
There is in fact very little ancient evidence both for the functioning
and for the disintegration of a comprehensive system of political clientela
in the Roman republic. It is essentially a modern hypothesis to explain
the domination of the elite, as observed not least in the assemblies. The
argument supposes that for the formal powers of the plebs to have been
so effectively neutralised each member must have been subjected to
tight personal control. This logic, however, relies on the assumption of a
`politicised' plebs, which would otherwise have acted as an independent
legislative body. That is a dubious supposition, not least in the light of
the structural limitations imposed on popular participation. These
limitations meant that the comitia could never represent `the Roman
people', only small sections of them. A tight, comprehensive system of
personal control is therefore not needed to explain the acquiescence of
the assembly before 133.
Something clearly happened in the later second century. Still, the
patronage model may not be the best tool to explain it by. We have no
compelling reasons to believe that the change was due to a more or less
stable group of popular representatives changing its behaviour in this
period. In this chapter I shall argue that it was the patterns of attendance
and perhaps also the level of attendance which changed, not the ties
and allegiances of the populace. Instead of a weakening of patronage we
may be dealing with the involvement of new groups which had pre-
viously remained outside politics. Considering the small scale of Roman
politics in general, that would seem a very realistic possibility. Thus, a
quantitative approach to the passing of `popular' laws and in particular
the senate's response may put the working of the assembly and the
underlying patterns of social control in a new light.
A long list of laws can be compiled which were passed against the
erce opposition of the senate. Ti. Gracchus' tribunate in 133 proved a
turning point. Admittedly the leges tabellariae of Gabinius and Cassius
and the Lex Calpurnia de repetundis had already been passed. Still,
Gracchus' programme was not only far more ambitious, including laws
on land distribution, the Attalos legacy, and involving the demotion of
his colleague Octavius; it also led to a change in the basic rules of
Roman politics. Ti. Gracchus left to his successors a range of political
tools and methods which had previously been unimaginable. Whether
by chance or intent he had pioneered a strategy which would be further
developed in the following generations. Thus, already Caius Gracchus
and his allies managed to implement an even more momentous pro-
gramme against the opposition of the senate, including a broad variety
of bills, among them agrarian, colonial, frumentary and judiciary. In the
following decades down to the late 50s agrarian bills were carried
against the senate's authority by Appuleius, 103, 100, Sex. Titius, 99,
P. Vatinius, 59; grain-distributions by Clodius, 58; judiciary laws by
Appuleius, C. Servilius Glaucia, 104/101, Clodius, 58; on citizenship
and tribal inscription by Sulpicius, 88, and C. Manilius, 67; voting by
Marius, 119, C. Coelius Caldus, 107, Cn. Domitius Ahenobarbus, 104,
(reinstated in 63 after its subsequent repeal). Sulpicius had Sulla's Asian
command abrogated in 88, and commands were conferred on Marius in
107 and 88, Pompey in 67; provinces were allocated to Caesar in 59 and
55 and to Gabinius in 58. Other laws included C. Cornelius' on the
praetorian edicts, 67, and Clodius' extension in 58 of the days that
could be used for comitia and his modication of the use of obnuntiatio.
Sulpicius also imposed limits on the debts of senators.
Many of these laws openly challenged vital interests of the senatorial
class, and were opposed with great though ineffectual determination.
The senate did manage to bring down a number of `popular' bills.
Generally, however, it concentrated its forces and energy on fronts other
than the voting itself, resorting to a number of alternative methods of
obstruction.
16
Tribunician intercession was an obvious weapon in the hands of the
senate. It worked on several occasions, e.g. in 122 against C. Gracchus'
citizenship bill; in 67 when C. Cornelius challenged the senate's right to
grant privileges; and in 63 the mere threat was effective against the
rogatio Servilia. The method was not foolproof, however. Occasionally
no willing tribune could be found. Moreover, a determined tribune
could have his colleague deposed, as happened in 133. Likewise, in 67
Gabinius was about to demote Trebellius when the latter chose to
withdraw his veto after seventeen tribes had voted for this demotion,
Asc. 72C; Dio 36.30. Moreover, a veto could simply be ignored, as
happened in 59, when Caesar's agrarian bill was forced through.
Legislation could also be prevented by the use of religious obstruc-
tion.
17
While a magistrate was observing the skies for omens no political
proceedings could take place. Declaring obnuntiatio thus automatically
stalled all activity in the comitia. The method was used against Caesar
and Clodius, who responded by imposing restrictions on the use of
obnuntiatio.
18
In 59 Caesar simply ignored his colleague's obnuntiatio.
Another strategy used with increasing frequency involved violence,
occasionally given a touch of legitimacy by the so-called senatus con-
sultum ultimum. By disrupting the assembly the senate could prevent a
16
See De Libero (1992).
17
See in general Taylor (1949) 7890.
18
Fezzi (1997).
vote from being taken. Senatorial leaders often turned up with bands of
strong-arm men in an attempt to obstruct the political process. Among
the best-known examples is the lex frumentaria of Appuleius Saturninus,
which was obstructed by the armed intervention of the senate, Rhet.
Her. 1.21.
Finally, if all other measures had failed, laws could be annulled by
senatorial decree, usually claiming an infringement of existing rules.
19
The most famous case comes from 91 when M. Livius Drusus' entire
legislation was declared null and void by the senate, using a number of
excuses. Later the laws of Sulpicius were annulled by Sulla. The lex
agraria of Sex. Titius, 99, was subsequently annulled, and an agrarian
bill, which M. Junius Brutus may have passed in 83, would later have
been rescinded by Sulla. In 67 C. Manilius' law to enrol freedmen in the
rural tribes was immediately repealed by the senate.
When looking at the various strategies used by the senate to counter
`popular' legislation one is struck by the complete absence of any serious
attempt to mobilise a popular crowd against the `populares'. Whenever
senatorial leaders turned up in the comitia with their bands of followers,
the intention seems to have been to disrupt the process rather than to
defeat the bills in the polls. We are left with the impression that the
senate was unable to put up any effective opposition in the comitia.
This situation has drawn little attention, probably due to the obvious
fact that `popular' laws in the nature of things are supposed to have
broad popular appeal. The senate, on this line of reasoning, would have
had little choice but to seek other means of obstructing hostile bills.
However, the capacity of the comitia allowed merely a small fraction of
the citizenry to participate. The senate was therefore not confronted
with crowds so huge they could not possibly have been taken on, let
alone outnumbered. The level of participation feasible in the comitia
puts the senate's inability to conduct an effective mobilisation in a new
perspective. In order to present a realistic counter-force only a modest
number of followers would have been required. Assuming for the
moment an absolute maximum of comitial attendance in the Forum of
around 10,000, the tables would have been turned if each senator had
mobilised just twenty-ve voters; and many fewer after Sulla had
doubled the number of senators. Most often the attendance would have
been considerably lower, as suggested, for example by the transfer of
voters to empty tribus, thus putting the senate's impotence into even
greater relief. Moreover, despite the very small numbers needed to put
up a credible opposition to the populares we hardly ever hear of over-
19
Heikkila (1993).
crowded assemblies which reached or exceeded the limits of the
assembly's capacity. In 167 Paullus' army, as we saw, lled the Capitol
completely. But otherwise there is no legislative assembly on record
which was too large for the venue, leaving voters struggling to get in.
20
The nobles' difculty in getting access to the Capitol in 133, for
example, was not caused by overcrowding but by deliberate obstruction
by Ti. Gracchus' followers.
In sum, therefore, the senate does not seem to have tried in earnest to
mobilise followers against the populares; from the rst emergence of this
opposition it relied on alternative methods of obstruction. The popular
followings, by which the elite had hitherto controlled the comitia so
effectively, would in other words have evaporated overnight as soon as
the rst populares appeared on the stage. Senatorial clients seem to have
disappeared from the political scene almost without a trace. They do not
feature prominently in any ancient source on late republican politics.
Mentioned only in passing, most often in the context of elections, we
rarely get the impression that particular importance was attached to this
category. Thus, when Cicero describes the warm reception Murena
received on his return to Rome, clientes are lumped together with groups
like vicini, tribules and the army of Lucullus, Mur. 69. Likewise, when
the senatorial forces used against the early populares are described, no
lower-class clients are mentioned. The participants are listed as sena-
tors, knights, friends and servants/slaves, `friends' being an unlikely
reference to working-class dependants, Plut. Ti. Grac. 18.2. In 100
Caepio blocked Saturninus' lex frumentaria `cum viris bonis', Rhet. Her.
1.21,
21
and later Marius also called up veterans from Picenum, Cic.
Rab. Perd. 22. Likewise, the senate used regular troops against C.
Gracchus in 121.
22
This all seems to suggest that the senate could
barely raise an effective guard against `popular' leaders out of their own
personal followings. Even when successful the senate's forces remained
quite small. Thus, in 133 they were outnumbered by Ti. Gracchus'
men. Later senatorial leaders formed guards, but even then we have
little evidence of the use of lower-class clients. As we have seen, Cicero's
guard was made up of knights and Italians from Reate.
23
Likewise,
20
The passing of the Lex Gabinia in 67 was, as we have seen, accompanied by a large
public demonstration of popular support for Pompey, but the sources also present the
event as highly exceptional.
21
Cf. Cic. Rab. perd. 23, mentioning ìlla armata multitudine bonorum', which had met
Saturninus in his nal stand against the senate and the boni.
22
Plut. C. Grac. 16.3; Oros. 5.12.7; Ampel. 26.2 (slaves).
23
Cic. Cat. 3.5; Phil. 2.16. According to Sall. Cat. 26.2, Cicero's guard was made up of
friends and clients. Still the social standing of the clients is not known. According to Val.
Max. 9.5.2 the consul Philippus was injured by a cliens of the tribune Livius Drusus in 91.
Cicero's reference to the `familiares' who assisted Curio in his violent
actions, Phil. 2.4, does not suggest they were humble clients. The only
reference to the employment of clients for political intimidation con-
cerns a clash between C. Cato's clients and his prosecutor Asinius
Pollio, Sen. Contr. 7.4.7. The mobilisation of household slaves rather
than clients seems to have been the natural alternative to calling on the
support of noble friends, which remained the primary option.
24
The absence of clients on the senatorial side raises questions of a
more general nature: for how could an apparently well-functioning
network of social control break down in such an abrupt and denitive
fashion? Vanderbroeck suggested that a section of the plebs, the ìnde-
pendent freedmen', had broken loose: à differentiation had occurred
within the Roman plebs. One group [i.e. the independent freedmen]
became detached from existing ties of patronage and sought new ways
for the articulation of demands. This void in the patronclient relation-
ship was lled in by popular leaders.'
25
This theory, however, carries
little conviction. Even if we accept that some plebeians broke away, that
leaves unexplained the absence of the remaining plebeians supposedly
still under elite control. Why were they not mobilised against the
intractable crowd of ìndependent freedmen'? The same objection
would apply against attempts to explain the populares in terms of
enfranchisement and immigration, ascribing the upheaval to an inux of
ìndependent' rustics who were not integrated into existing systems of
patronage. For again, this model would not account for the stance taken
by the old clients who seem to have completely deserted their patrons in
their struggle with the populares.
The logical conclusion to the story of the missing clients would be to
abandon the idea of political clientela as the arcana imperii of the
republican elite. There is simply no evidence to suggest that the
supremacy of the senate had ever been based on personal patronage.
Since the large majority of the plebs could never have attended the
assembly anyway, it might be more fruitful to operate with politically
active and passive sections of the population rather than trying to
explain the `popularis' phenomenon in terms of `controlled' and ìnde-
pendent' crowds.
The assumed decline of clientela was, as we have seen, derived from
the political changes in the late republic, but there are many indications
that the senate's unrivalled authority in previous periods was not based
24
In Phil. 2.16, Cicero assured his audience that his guard in 63 had been knights and
good citizens, and not, as it was alleged, armed slaves, which presumably was common
practice, cf. Nowak (1973) 78.
25
on personal control over plebeian voters. Thus, several laws opposed by
the senate had been passed long before the emergence of the populares in
133. The most conspicuous example is Flaminius' legislation in the late
third century. His Lex agraria from 232 was passed by the concilium
plebis, for the rst time making use of the Lex Hortensia, which granted
law status to plebiscita. The bill met with the senate's disapproval and
was unable to get through the comitia centuriata, Cic. Inv. 2.52. Also
later there are signs that the comitia could not be relied upon to follow
the senate's recommendations on every issue. That seems, for example,
to be the implication of the row in 188 over the assembly's right to grant
suffragium to the Volscian towns of Formiae, Fundi and Arpinum, Livy
38.36.78. The rogatio of the tribune Valerius Tappo was vetoed by four
of his colleagues on the grounds that it did not have the senate's
approval. They were then told that the people, and not the senate, had
the right to grant full citizenship. The course of events suggests that the
comitia were not completely in the hands of the senate but could pass
bills on their own. Likewise in 189 or 188 the tribune Q. Terentius
Culleo passed a law on tribal registration without senatorial sanction. As
Plutarch, Flam. 18.1, says, Culleo `wanted to spite the nobility and so
persuaded the people to vote for the measure'. In 149 the afore-
mentioned inquiry into the conduct of Galba in Spain, proposed by the
senate, failed to pass the assembly. Later in 139 the Lex Gabinia
tabellaria introduced the secret ballot in the election of magistrates, and
a further extension followed in 137. Similarly, the rst quaestio de
repetundis, prosecuting senators for extortion, was set up by the Lex
Calpurnia in 149.
It follows from these examples that it had also been possible earlier to
overcome senatorial opposition in the comitia. Instances may have
become more frequent in the latter half of the second century, but since
the comitia were also able to act independently prior to this period, no
general decline in social control can be inferred from this.
There can be little doubt that clientela was a signicant feature of
Roman society. As a social practice patronage pervaded all aspects of
private and public life, which relied on personal relations negotiated
through the constant exchange of favours and obligations. That,
however, does not necessarily imply that we are also dealing with a
socio-political structure which could form the basis for long-term
political domination by a ruling nobility.
Objections to this idea have been raised increasingly in recent years.
26
26
Millar (1984); Develin (1985) 12731, 3258; Brunt (1988) 2732, 382442 esp.
41424; Wallace-Hadrill (1989) 70; Gruen (1991) 253; Nippel (1995) 33; Pani (1997)
1978; Morstein-Marx (1998).
Millar has summarised the criticism in three points: (1) the size of the
city's population, (2) the internal competition within the elite, (3) the
importance of rhetoric in the political process.
27
The size of the popula-
tion is crucial. As Wallace-Hadrill noted, the plebs was: `too numerous to
enter into signicant personal relations with the few hundred members
of the political elite'.
28
It does indeed seem unrealistic that the entire
population could have been tightly organised into stable vertical struc-
tures of dependence, each headed by a senatorial family. Moreover,
since the city of Rome, simply to maintain its population, had to rely on
a constant ow of immigration, the urban population was not only too
large but also too uid to t this rigid model of personal control.
The importance of rhetoric, and the political independence of the
audience which it presupposes, is more dubious. The argument forms
part of Millar's democratic interpretation of Roman politics, which, as
we have seen, pays little attention to the practical aspects of popular
participation. It is, however, precisely these issues which must be
considered if we are to understand the political impact of rhetoric. The
size, composition and motivation of the audience become all-important:
the rabble-rousing effect of popular oratory on a crowd of followers
should not be confused with the power of argument.
29
The elite's internal competition over public honours, inuence and
clients meant that political allegiances could never have been as clear-
cut as is often assumed. This highlights an important aspect of clientela,
one which has been brought out more clearly in modern studies of the
Roman elite. Here traditional concepts of clientela as a stable structure
are called into doubt, as historians start questioning the exclusive and
long-term character ascribed to these relationships. A new consensus is
developing around the view that patronage did not determine political
groupings and alliances, which all appear to have been short-lived and
generally focused on specic issues.
30
The old model of permanent
family factions, often based on inherited patronage, will have to be
27
Millar (1984) 2, 10.
28
Wallace-Hadrill (1989) 69.
29
Millar (1998) 1505 argues, for example, that rhetorical persuasion was crucial even
before the vote to recall Cicero from exile in 57. On this occasion, however, the
assembly was lled with boni and Italian municipales, who had been mobilised for this
particular purpose. It therefore seems highly unlikely that they would have considered
opposing the bill. The speeches delivered by leading senators were part of a public
demonstration of unity between the senate and the boni. As such it may also have
served as a spectacle put on for the benet of the municipales who would rarely have had
the experience of seeing the legislative process at work.
30
Fundamental is Meier (1980); see also Develin (1985) and Brunt (1988). The
traditional position has recently been restated by Briscoe (1992).
abandoned; there are few, is any, certain traces of such long-term
alliances in Roman politics.
These results tie in well with recent anthropological research which
suggests that clientela is not typically an exclusive or permanent relation-
ship.
31
In most societies clientela has been a uid and competitive
system, characterised by pluralism, choice and instability. The overall
impression seems to be one of ambiguity and exibility rather than
rigidity and permanence. In Rome these qualities may have been
obscured by a public ideology which glossed over internal contradiction
in order to present a simple unied structure. They may, however, have
surfaced in the social practices associated with clientela, and in particular
the salutatio, the ritualised morning reception where clients turned up at
their patron's house. Thus, Cicero, Mur. 44, spoke of `this new practice
of running in a crowd from house to house',
32
and in the Commentar-
iolum we are told that: `The callers are a more promiscuous crowd, and
in the fashion of today visit more than one candidate.'
33
These refer-
ences have been cited in support of the theory of a decline of clientela in
this period.
34
It is more likely, however, that they simply reect the
inherently uid nature of patronage. The habit of visiting more houses,
here specically linked to electioneering, may have been new, but it
probably reects an ongoing negotiation which may always have been
present in clientpatron relationships. Thus, the Commentariolum goes
on to describe how candidates would compete over the loyalty of
potential clients by dispensing attery and attention.
The introduction of the secret ballot in the second century has been
seen as a factor in the decline of patronal authority, while the senate's
opposition to it is taken as an indication of its previous importance as a
means of controlling the comitia. Cicero deplored the reform, claiming
that the secret ballot had freed the people who had previously been
òppressus dominatu ac potentia principum', Leg. 3.34, cf. Sest. 103.
Notably, Cicero does not explicitly talk about clients; he simply con-
trasts the people with the principes, who were no longer able to check
their voting in public. Moreover, we have no indication that the secret
ballot had any inuence on the emergence of `popular' legislation.
35
Its
primary importance lay in elections, where it was rst introduced for
31
Johnson and Dandeker (1989).
32
`. . . hoc novo more omnes fere domos omnium concursent . . .'
33
Comm. pet. 35: Ìn salutatoribus, qui magis vulgares sunt et hac consuetudine quae
nunc est [ad] pluris veniunt . . .'
34
E.g. Vanderbroeck (1987) 1045, who identied these promiscuous clients as
independent freedmen and members of the plebs contionalis.
35
Gruen (1991) 259.
reasons which now escape us;
36
in 119, Marius narrowed the pontes to
ensure further secrecy, again probably with a view to the elections.
37
Two years later, in 137, the secret ballot was also introduced in trials;
but it was not until 131 that the new principle was applied to the
legislative comitia. It follows that Ti. Gracchus' laws had all been passed
by oral voting, as had the rst leges tabellariae and the Lex Calpurnia de
repetundis. Ti. Gracchus appears to have made no attempt to introduce
the secret ballot here, suggesting that it was not seen as in any way vital
to the promotion of `popular' legislation.
One possible explanation might be that in order to pass their bills the
populares relied on sections of the population which were not integrated
into elite patronage. For it is not a given fact that clientela always formed
a direct link between the top and bottom of society. The social and
economic distance between the political class and the majority of the
population was overwhelming. Doubts have rightly been expressed as to
the comprehensive nature of clientela. Cloud noted that: `Rome, like
other societies, excluded the very poor from the clientpatron relation-
ship.'
38
And according to Tacitus, Hist. 1.4.3, the better part of the plebs
was linked to noble houses, the implication being that the plebs sordida
was not.
Thus, while there is no doubt that Cicero and other senators had
clients, the question is to which social stratum they belonged. We cannot
automatically assume that they would have included members of the
working population. Thus, when Cicero for example, in the context of
consular elections, mentions `homines tenues' among a candidate's
active supporters, Mur. 70, he may not refer strictly to the `poor' in the
literal sense of the word. Poverty is always relative, and Cicero obviously
36
Important attempts to explain the secret ballot have been made by Hall (1990) 1967,
and Jehne (1993). The traditional view of the Lex Gabinia as a `popular' attempt to give
the people greater freedom in elections has been widely abandoned, although Marshall
(1997) has recently restated this position. Some now prefer to see it as a measure
against bribery, curbed by the inability to check individual votes cast by ballot, e.g.
Thommen (1989) 823, Lintott (1990) 7. According to Hall it was an attempt by new
families to break the dominance of the old elite which had controlled the elections
through their patronal powers. Against this theory Jehne has argued that the records
show no increase in new men reaching ofce after the introduction of the secret ballot.
His own explanation is ingenious but ultimately unconvincing. It suggests that the aim
was to cover up `Loyalitatskonikten' caused by multiple patronal links and obligations.
Publicly exposed as they were by the oral vote, they threatened to bring down the whole
ideology of patronage. However, if the point was to save the face of clientela as a unied
and unambiguous system of loyalties, the bill would have been counter-productive. The
passing of a law granting secrecy to the clients' vote would have been the most blatant
admission of failure, more effectively undermining the system than any occasionally
surfacing conict of interest.
37
Cic. Leg. 3.38, cf. Marshall (1997) and Crawford (1974) 3067.
38
Cloud (1989) 210.
approached the issue from a senatorial viewpoint.
39
Members of `the
great unwashed' are unlikely to have been appreciated in the personal
following of consular candidates. In the electoral campaign for the
censorship in 142 Scipio Aemilianus had been criticised for having
people `who were of low birth and had recently been slaves' in his
entourage; two individuals were singled out: a herald, who was probably
one of his own freedmen, and a freed publicanus.
40
The `tenuiorum amicorum et non occupatorum', mentioned by
Cicero, were friends who joined a candidate's entourage because that
was the only way they could pay their duties to him. But that simply
means that they could lend no direct political favours, that is, they were
not themselves members of the political class; only by their personal
attendance could they show their respect. However, their sheer ability to
do so would suggest that rather than being members of the working
classes, they were men of means who could dispose freely of their own
time.
41
The case against clientela as a signicant political factor, especially
outside the propertied classes, is strongly reinforced by the logistics of
late republican politics. The size of the popular institutions calls into
question the idea of a comprehensive system of political clientela. At
least until 145 the legislative assemblies, that is, those held in the old
Comitium, had numbered fewer perhaps considerably fewer than
3,600 people. This level of attendance is not easily reconcilable with a
model of general control and political exploitation of the entire citizen
population. We are dealing with two radically different orders of magni-
tude: on the one hand, the number of clients in Rome who would be
expected to provide political services and, on the other hand, the very
limited turnout which was practically possible. Between the two there
39
Cicero, for example, describes an aedilician scriba, D. Matrinius, as a `homo tenuis',
Cluent. 126, despite the fact that scribae held a relatively high status, forming an ordo of
their own, cf. Badian (1989).
40
Plut. Aem. Paull. 38.34; Mor. 810B.
41
Vanderbroeck (1987) 83, describing them as `poor people who had nothing else to do',
seems to interpret `non occupatorum' as ùnemployed'. But Cicero simply contrasts the
senators and equites, who only had time to pay shorter visits and perhaps accompany
candidates to the Forum (cf. Cic. Att. 2.1.5), with those lower ranking friends who
could take the whole day off from other duties to follow their candidate. The evidence
of Cicero's letters gives no hint that `poor' people were involved in his morning
receptions or present in the crowds which followed him to the Forum. Thus, in Fam.
7.28.2 he speaks of `salutatio amicorum', and in Att. 1.18.1 the sectatores, who had rst
lled his house at the morning reception (completa domus), were clearly men of
considerable social standing. Fam. 9.20.3 suggests that the salutatio was performed
exclusively by people with an active hand in politics, falling into the categories of either
boni or Caesarians. Likewise the Catilinarians who had turned up for Cicero's salutatio
with the intention of assassinating him were senators and equites, Sall. Cat. 28.1.
seems to be no reasonable balance which could justify the maintenance
of a comprehensive system of political control.
Any study of the relationship between politicians and voters must
start from the fact that an assembly was the gathering of a very small
section of the citizenry. As argued above, the notion of regular participa-
tion by a sub-group within the working classes, the plebs contionalis,
seems implausible; the natural participants were men of means who,
given the capacity of the assembly, could easily have lled the popular
institutions on their own. The make-up of the crowd was mirrored in
the social bias built into the assembly; the system deliberately favoured
the landowners who lived in Rome, at the expense of the lower urban
classes.
As we have seen above, Diodorus' description of Octavius' followers
in 133 suggested that the traditional political crowd in Rome were well-
to-do and naturally aligned themselves with the senate against a
`popular' threat. Occasionally they were supplemented by voters who
came in from the countryside, as happened in 195 when the anti-luxury
law, the Lex Oppia, was repealed.
42
Again, these voters clearly did not
belong to the lower classes, who had neither the opportunity nor in
this case the incentive to travel to Rome. The motives behind their
participation can only be conjectured. However, since proposals were
hardly ever rejected, it could rarely have been driven by a personal
interest in particular issues. Most people probably came out of habit and
for the prestige associated with the exercise of political rights essen-
tially a gentleman's pursuit. The sheer entertainment value should not
be underestimated either.
Many participants would have enjoyed noble patronage. But whether
this was signicant for their voting is uncertain. The uid and non-
exclusive character of clientela would frequently have created conicts of
loyalty. When the elite was internally split, considerable scope was left
for individual choice. Viewed in that perspective it is not inconceivable
that on some occasions public debates might have had a real impact on
the outcome. One such case was the debate over the inquiry into
Galba's conduct in 149. Here his emotional plea was able to sway the
assembly against the inquiry, which had been proposed by the senate.
The implications of such instances should not be exaggerated, however;
obviously no vital interests were at stake for the elite on this minor issue.
Cicero unsurprisingly elevated the role of oratory, claiming that
rhetoric on several occasions had prevented `popular' legislation, for
example the Leges Licinia and Papiria. But since the debate at pre-
42
Livy 34.1.6: èx oppidis conciliabulisque conveniebant'.
comitial contiones in the nature of things could inuence only those few
who had decided to turn up, these defeats are probably better explained
by a failure to mobilise popular supporters.
To summarise, before the rise of the populares legislative comitia may
by and large have been a constitutional formality, attended by small
groups of probably well-to-do citizens who only very rarely exercised
the choice put before them, thus allowing the authority of the senate to
prevail almost invariably. Many of the participants may have had
personal links to the political class, but their compliance was most likely
due to the overall community of interests enjoyed by the propertied
classes. The comitia on this interpretation therefore appear as an exten-
sion of the political class itself rather than a separate institution
representing the people. It follows that in this period there was no
political counterweight to the rule of the senate.
When this cosy arrangement broke down in the later second century it
was probably not because the traditional participants in the comitia had
for some reason turned against the senate. More likely it was a conse-
quence of members of the lower classes now turning up for assemblies
they had not previously attended. That happened at the initiative of
magistrates who sought popular support to press through legislation
against the opposition of the senate and the upper classes.
Mobilisation of these voters was a novelty; and one which left the elite
with a sense of disgraceful anomaly. Their contempt for the plebs sordida
exercising their citizens' rights surfaces in Scipio's stance at Carbo's
contio in 131, and later in Cicero's indignant dismissal of the `poor', who
supported Clodius. It seems that in aristocratic Roman society politics
was naturally considered the dignied preserve of men of means.
An early example of lower-class mobilisation secured the passing of
Flaminius' Lex agraria in 232, a law which had had no chance in the
centuriate assembly. Much later the Leges tabellariae were passed against
the authority of the senate, although it remains uncertain whether that
happened through popular mobilisation. The Lex Gabinia might have
been passed in a traditional assembly by sections of the upper classes,
hoping to break the nobility's hold over the elections.
43
This law was
followed by the Lex Cassia, over which the elite appears to have been
split; according to Cicero, Scipio Aemilianus recommended the bill,
Leg. 3.37. These bills may be signs of the growing disunity within the
elite, which fully erupted in 133, when for the rst time since Flaminius
popular mobilisation was organised on a serious scale. The strategy was
new and untried, however, and the defeats of Papirius and Philippus,
43
Cf. Hall (1990).
and the weakening of C. Gracchus' position during 122 all suggest that
popular mobilisation was not yet fully reliable as a political tool. Later
the `popular' machine became more efcient, culminating in Clodius'
organisation of the plebs. In the rst century no `popular' leader suffered
legislative defeats against the authority of the senate.
The corporate structure of the assemblies would have been a signi-
cant factor in the mobilisation of popular support. The different
weighting of electoral groups meant that for a crowd to be effective it
would have to be raised with considerable care and circumspection. The
distinction between rural and urban tribus was crucial; the inscription of
the urban plebs in only four tribus had created a paradox. For while rural
voters had great potential inuence but little chance of exercising it,
urban voters had easier access to assemblies but much reduced inu-
ence. That naturally left the comitia in the hands of urban landowning
citizens inscribed in the rural tribus.
Viewed in this perspective the importance of immigration becomes
obvious. In the second century the city of Rome grew substantially, as
underlined for example by the need to build new aqueducts during the
second half. This increase must be set against the relatively higher
mortality rate in the metropolis, which meant that simply to keep its
population constant a permanent supply of manpower was required.
44
The two main sources were the import of slaves, later to be enfran-
chised, and immigration, primarily from the Italian peninsula. The ratio
between these two factors cannot be determined. Still, immigration
from Italy may have reached a considerable level in this period. There
are many signs of a ow of people from the interior regions to the highly
urbanised areas along the Tyrrhenian coast and further to Rome itself.
The immigration of Italian allies who had no claim to Roman citizenship
obviously attracted most attention. Thus, foreigners were expelled in
124 by the bill of Pennus, while illegal usurpation of Roman citizenship
by immigrants was prosecuted by the Lex Licinia Mucia in 95.
45
Most
likely the inux of Roman citizens was even greater than that of
foreigners, who faced a precarious existence with no rights and threat-
ened by expulsion.
Thus, the number of rural tribules living in Rome may plausibly have
multiplied in the late republic. No matter which view is taken of the
extent of previous patronage, this development would have had signi-
44
Thus recently Scheidel (1994) who concludes that: `. . . a constant inux of immigrants
would have been necessary to compensate for the effects of tuberculosis, hyperendemic
malaria and other infectious diseases . . .', 166. See also Morley (1996) 4454. Purcell
(1994) 6578 takes this argument too far, exaggerating the uidity of the urban
population.
45
Cic. Off. 3.47; Balb. 54; Brut. 63.
cant consequences in terms of social control and civic cohesion. While
the other main category of new urban citizens, the freedmen, generally
maintained more or less close links to their former masters, the immi-
grants probably remained outside existing social networks for some
time. We are therefore faced with yet another paradox of late republican
politics: that the most marginal group in terms of social control was able
to wield much greater inuence in the assembly than the one most
closely tied to the elite. During the rst century the number of immi-
grants grew even further, boosted not least by the enfranchisement of
the former Italian allies in the 80s.
46
Their tribal inscription was not
remotely effective until 70, but those already living in Rome may have
been covered by the limited registration carried out in 87.
The question remains of whether the emergence of the populares can
be traced directly back to this development within the demographic
basis. It is indeed tempting to see the immigrants' political power, low
social status and relative independence as primary factors responsible
for the collapse of the senate's authority in the legislative comitia. That
explanation, however, runs the risks of monocausality. For although the
numbers increased in the second century, there may always have been
poor rural tribesmen living in Rome. Considering the small numbers
involved in the passing of legislation, it would probably also have been
possible to organise a sufcient turnout of rural tribesmen before the
late second century crisis.
47
In the early `popular' phase some efforts were made to mobilise voters
from the rural tribes in the countryside outside Rome. Thus, in 133
rustici helped Ti. Gracchus depose Octavius and also passed his agrarian
reform.
48
Later they were again solicited when Ti. Gracchus prepared
for what was to be his last assembly.
49
Also C. Gracchus used rustici to
pass his legislation and campaigned in the countryside, delivering a
speech `de Popilio Laenate circum conciliabula'.
50
However, this type of
mobilisation soon turned out to be unreliable, laborious and ultimately
46
Sall. Cat. 37.78, refers to the immigration of unruly elements from the countryside.
47
In principle immigrants would have been transferred to the urban tribus at the rst
available census. Whether that always happened in practice may be doubted. Many
may have tried to avoid transfer to the low-status urban tribus, and for up to ve years
they would in any case have maintained their rural afliation. The census was,
moreover, highly irregular in the rst century, the census of 70 being the last to be
conducted under the republic. Brunt (1966) 67 believed many plebeians to have been
inscribed in the rural tribus, cf. Last (1932) 79; Millar (1998) 16. Purcell (1994)
6578; (1996) 797, denies the existence of poor ingenui in Rome, making the curious
claim that the entire urban plebs consisted of freedmen, cf. Millar (1998) 36.
48
Diod. 34/35.6.1; App. B. Civ. 1.567.
49
App. B. Civ. 1.589.
50
Cic. Cat. 4.4; Plut. C. Grac. 3.1; 13.2; Gell. 1.7.7.
superuous. It was simply not possible for rustici to maintain a constant
political presence in Rome, and dispersed as they were they were also
more difcult to address and mobilise than urban voters. Their absence
from Ti. Gracchus' last assembly due to the harvest illustrates the
problems involved. In that situation Ti. Gracchus had to improvise and
quickly raise support among the urban plebs. Nevertheless, it seems that
after a day's delay he was gaining the upper hand, when the senators
intervened, App. B. Civ. 1.60. The early attempts at rural mobilisation
do not imply, as has been claimed, that rural voters still represented the
majority in the assemblies, nor that members of the urban plebs were
tied to their patrons.
51
The Gracchi probably campaigned outside Rome
because the strategy of popular mobilisation was new and virtually
untested. The importance of the rural tribes was obvious to any Roman
legislator; basic political instincts would have told the rst populares to
campaign among the poor rustics. However, it emerged in the process
that there might be sufcient numbers of urban plebeians inscribed in
the rural tribes to carry legislation through on their support alone.
Thus, Appian, 1.603, also refers to Gracchus' poor urban supporters,
and we have no indication that C. Gracchus' laws, or the Lex Papiria
tabellaria for that matter, were all passed by rustici. Apparently none of
the later `popular' leaders saw any reason to campaign outside Rome.
Saturninus' use of agrestes seems to have been atypical; the rural voters
turning up in Rome were apparently veterans of Marius and prospective
beneciaries of the bills, Plut. Mar. 28.5. Later there appears to have
been little need for rural voters. Increased immigration to Rome during
the second century had already created a solid base of urban voters with
an effective say in the comitia tributa and the concilium plebis. Once
`discovered' this resource was there to be exploited by a succession of
`popular' leaders.
We know very little about the way in which the Gracchi and other
early populares in practice raised their crowds. But it does seem that in
this period popular mobilisation was still in its infancy and its func-
tioning somewhat erratic. Thus, at Ti. Gracchus' rst attempt to pass
the bill to allow tribunician iteration there was an insufcient turnout,
and C. Gracchus' failed second re-election, albeit on grounds of manip-
ulation, also suggests agging voter support. The defeats of Papirius and
Philippus in 130 and 104 point in the same direction, and also the near-
defeat of one of C. Gracchus' bills, probably his judiciary reform, Diod.
34/35.27. Appian, B. Civ. 1.59, describes how Tiberius approached
51
For the former view see Nicolet (1977) 356, for the latter Nippel (1981) 77. Taylor
(1966a) 667 believed the Gracchi mobilised outside Rome, because the rural
tribesmen in Rome all belonged to the propertied classes.
people at random in the Forum asking for their help in passing his last
bill. As we saw, the attempts by the Gracchi to mobilise rustics also
suggest that effective political networks had yet to be established among
the urban plebs.
This conclusion might seem to be contradicted by Plutarch's descrip-
tion of the low-ranking men in Scipio Aemilianus' entourage who were
`frequenters of the Forum and able to gather a mob and for all issues by
means of solicitations and shouting', Aem. Paull. 38.34. The story,
however, is confused and ts badly into the context, which is the
censorial election of 142. To make proper sense the people in question
would have been electoral xers in the timocratic comitia centuriata, and
not organisers of popular crowds forcing bills through the tribal as-
sembly; in this period we have no trace of such middlemen or regular
use of popular crowds in legislation.
When popular mobilisation took off in the rst century the vici seem
to have become the basic unit.
52
They were the smallest unit in the
urban divisions, focused on the cult of the compitales. Each neighbour-
hood had its own shrine and local organisation whose ofcials celebrated
the annual festival of the compitales. The vicinal collegia may have been
closely linked to the professional associations; they may even have over-
lapped, trades often being concentrated in particular streets and areas.
53
The vici rst emerged as a political resource in the 80s, when M. Marius
Gratidianus appears to have cultivated them.
54
Their activity in the 60s
is indicated by the ban on collegia in 64(?), which also entailed the
discontinuation of the Compitalicia, Asc. 7C; 75C.
55
But it is not until
the time of Clodius that we get a somewhat clearer picture of the
methods used in this type of mobilisation. In 58 he again allowed the
Compital festival to be celebrated, putting his aide Cloelius in charge,
Cic. Pis. 8. He also passed a bill òn the restoration of collegia and the
establishment of new ones', `de collegiis restituendis novisque insti-
tuendis', Asc. 7C, cf. Cic. Pis. 9; Sest. 55. The local character of these new
political clubs is indicated in Cicero's references to Clodius' mobilisation
being conducted `vicatim', Sest. 34, and his followers as èx omnibus vicis
concitatos', Dom. 54. His local clubs were divided into smaller units,
decuriae, which was a common form of collegial organisation.
56
And as
52
On the political activities of the vici see (with caution) Flambard (1981), which has
now been superseded by Fraschetti (1990) 192250.
53
Cf. Morel (1987) 1434; Coarelli (1984) 467.
54
Statues of him were erected ìn omnibus vicis', Cic. Off. 3.80; Sen. De ira 3.18.1; Pliny
NH 33.132; 34.27.
55
The date is not entirely certain; Fraschetti (1990) 21617, puts the ban in the 60s.
56
Cic. Sest. 34: `cum vicatim homines conscriberentur, decuriarentur'; Post red. Quir. 13;
Dom. 13.
the Commentariolum suggests, the ofcials of such organisations appear
to have played a central role in the mobilisation of their members.
Cicero is advised to gain the support of the principes of collegia, montes,
pagi and vicinitates, since: `through them you can easily secure the
masses that remain'.
57
For the crowds to be effective in the assemblies the rural tribesmen
who lived in Rome had to be specically targeted. How this was done
must remain a matter of conjecture. `Popular' leaders may have tried to
establish contacts with the tribal organisations, which had their own
headquarters and ofcials.
58
We know that magisterial candidates
courted the tribal organisations; the possibility exists that the populares
may have made similar overtures. Otherwise the collegia may have offered
an opportunity to attract the rural tribesmen in Rome. Thus, Patterson
has linked the growing importance of collegia in the rst century with the
inux of immigrants to Rome.
59
In the history of large cities associations
of this type have often provided the initial social support network for
newly arrived provincials. In that perspective it may not be too far-
fetched to assume that immigrants to Rome would have been particularly
attracted to Clodius' political clubs and warmly welcomed there.
There has been a tendency among scholars to perceive the following
of `popular' leaders as a form of public clientela.
60
It is assumed that the
populares had established a personalised relationship with plebeians
otherwise deprived of noble patronage. That, however, means a further
extension of a model which has itself come under increasing attack in
recent years. Obviously, its application to the populares must also be
questioned. Most `popular' leaders had followers who were bound to
them by a strong sense of personal loyalty. Ti. Gracchi allegedly never
went out with fewer than 3,0004,000 followers, many of whom stayed
with him until the bitter end. Loyal supporters also surrounded Satur-
ninus during his nal showdown with the senate. Clodius commanded
an effective force of devoted plebeians, who turned up for demonstra-
57
Comm. pet. 30: `. . . urbis totius, conlegiorum, montium, pagorum, vicinitatum, ex his
principes ad amicitiam tuam si adiunxeris per eos reliquam multitudinem facile
tenebis'. Fraschetti (1990) 20468 has argued convincingly that the magistri vici were
an Augustan invention, linked to the reorganisation of the regiones and the new
association of imperial and compital cult. He therefore sees the republican vici simply
as neighbourhoods with no formal structure or ofcials. However, the formulation
found in the Commentariolum suggests that the local neighbourhoods had some
recognisable leadership, which could be identied and approached by electoral
candidates. The vicinitates mentioned in Cic. Mur. 47: `. . . homines honesti atque in
suis vicinitatibus et municipiis gratiosi . . .', probably refer to the countryside rather
than the city of Rome.
58
Taylor (1966a) 69.
59
Patterson (1994) 237.
60
E.g. Vanderbroeck (1987) 81.
tions whenever called upon. But it is uncertain whether these groups are
best understood in terms of clientela. The case of the Gracchi, in
particular, seems no different from many other popular charismatic
gures who have throughout history attracted dedicated bands of
followers. There appears to be little to gain from forcing their personal
popularity into a general model of clientela.
Moreover, these supporters should probably be at least partly
distinguished from the crowds who passed the `popular' bills in the
assembly. A `rapid reaction force' may have been relatively easy to
mobilise among the plebs, especially if economic compensation was
offered. But however useful freedmen and other urban tribesmen may
have been in lling the contiones, they were much less effective in the
comitia, where tribal allocation was all-important.
The `popular' voters probably belonged to a different category of
followers, although of course some overlap cannot be ruled out. Their
support could not be taken for granted, nor does it seem to have been
based on any sense of personal or political obligation. Their inconsis-
tency is well illustrated by the power struggle between the elder Livius
Drusus and C. Gracchus, the former easily gaining popular support for
his blatant attempts to outbid his `popular' opponent, whose position
was so effectively undermined that his attempt to win a second re-
election failed. It seems that when a proposal appeared to be benecial
to the plebs, the intentions behind it were deemed irrelevant.
No popularis enjoyed blanket popular support; bills might fail or have
difculty passing despite their sponsors' personal followings and
`popular' credentials. Thus, some of their proposals clearly did not
strike a chord with the plebs, and were therefore never put before the
assembly, for example Flaccus' enfranchisement bill from 125. Also the
citizenship bill of the younger Drusus, proposed in 91, seems to have
lacked popular support, despite his extensive `popular' bribe of the plebs.
Likewise, some judiciary laws may have had less popular appeal,
explaining why C. Gracchus' reform of the judiciary seems to have just
scraped through.
Other `popular' laws had great difculty getting through in a regular
fashion. Cornelius' law on the libertine vote, an issue in which the
freeborn plebs had little interest, was carried at an assembly which had
not been properly called. The passing seems to have been something of
a coup and was subsequently annulled by the senate. Likewise, Clodius
was unable to prevent Pompey from taking charge of the annona in 57, a
commission he had already entrusted to his friend Cloelius.
61
Apart
61
Cic. Att. 4.1.6f, cf. Flambard (1977) 151; Garnsey (1988) 216.
from underlining the distinction between Clodius' personal followers
and his voters, this failure also shows that despite his unprecedented
level of popular organisation even Clodius had to present his case to his
constituency on every single occasion.
The plebs, it seems, viewed bills individually and largely in isolation
from their political context. The younger Drusus, though hardly a
popularis himself, certainly took a leaf out of their book to promote the
cause of the optimates. Land and grain bills were introduced to pave the
way for a judiciary reform, which would restore the senate's control over
the repetundae court. But it seems that for the bribes to be effective in
promoting the presumably less popular judiciary bill, the measures
had to be very closely linked. Thus, Cicero, Dom. 50, indicates that they
were combined in a way that laid Drusus open to accusations of
legislating `per saturam'.
62
In other words, even after a display of almost
unprecedented generosity, the plebs could not be relied upon to pass his
judiciary reform in a separate bill. This was therefore tied directly to the
more appealing `popular' measures. Each bill thus had to be commu-
nicated persuasively to the plebs. This was done informally through
assistants and local networks, although the contio also remained a focal
point in this process. The stakes were high; if the mobilisation failed the
voters might not be there on polling day. Therefore, the main problem
in passing a controversial bill was probably to persuade ordinary
Romans to turn up in the comitia.
When Gruen writes that Clodius `successfully mobilised a large
segment of the plebs with an attractive program that short-circuited the
usual lines between humble clients and their noble patrons', the impli-
cation is that they would otherwise have voted with their patrons.
63
But
populares do not appear to have struggled to break elite control over the
electorate; there is no trace of such a conict over popular loyalties. It is
more likely that the voters would not have turned up at all. For it should
be borne in mind that Gruen's `large segment' in any case could not
have included more than 34 per cent of the urban citizenry. The real
challenge to `popular' legislators lay in breaking the traditional political
passivity of plebeian voters.
That would help to explain the packaging of Drusus' legislation in 91.
The plebeians would probably not have directly opposed his judiciary
bill, but since they had little interest in the composition of jurors' panels,
he could not be sure they would be there for the vote. Likewise, Clodius'
traditional supporters are unlikely to have turned against him on the
issue of the grain commission. But with the free dole secured and
62
Cf. Ryan (1994) 1078.
63
Gruen (1974) 438.
Pompey known as a safe pair of hands they had no real incentive to turn
up in large numbers. While senatorial attacks were effectively neutra-
lised, competing `popular' bills might be more tricky to fend off.
It seems evident from these instances that the plebs had not gained a
constant presence or inuence in the assembly after the supposed
weakening of elite patronage. The unstable nature of popular participa-
tion suggests that we are not dealing with a clear-cut transition from a
controlled to an independent, active plebs; a point further underlined by
the anti-`popular' laws which could still be passed after the Gracchan
watershed in Roman politics.
After the death of C. Gracchus his murderer P. Popilius Laenas was
recalled from exile by an assembly in 121/120; in the following years
also the agrarian reforms were gradually dismantled. Likewise, the Lex
Servilia of 106 restored some control to the senate over the repetundae
court, which C. Gracchus had entrusted to the knights. In 104, as we
have seen, Philippus' agrarian bill was rejected, and in 99/98
Q. Caecilius Metellus Numidicus, the nemesis of Saturninus, was
recalled from his exile. Within the same decade C. Gracchus' Lex
frumentaria was modied by the Lex Octavia, probably reducing the
number of grain recipients. In 98 the Lex Caecilia Didia limited the
ability of populares to carry legislation. And even during the most
ruthless of senatorial restorations Sulla had laws passed in the comitia.
64
This pattern of alternating `popular' and senatorial legislation strongly
suggests that no regular plebs contionalis or legislative crowd existed. An
independent crowd, or in any case one which could now hide behind the
secrecy of the written ballot, is unlikely to have rst supported the
Gracchi, Saturninus and Sulpicius, and then endorsed the senatorial
backlashes following their defeats. We may be dealing with different
crowds, backing different policies. Quite plausibly, the plebs was simply
not present when the senate struck back. Mobilisation was crucial to the
assertion of popular control over legislation; without leadership from
members of the senatorial elite the people were effectively powerless.
When leadership was absent, and only when it was absent, as in the
aftermath of the senatorial showdowns with `popular' champions, could
anti-`popular' laws be passed. That happened, as we have seen, after the
crushing of C. Gracchus' and Saturninus' reform attempts, and when
the Sullan victory in the civil war had eradicated all opposition.
When, on the other hand, popular leadership was effective, the people
were able to rule supreme. Then the senate had virtually no chance of
putting up any real opposition in the assembly. For example, Minucius'
64
Gell. 2.24.11; App. B. Civ. 1.468.
attempt to repeal the Lex Rubria and abolish the Gracchan colony
Junonia seems to have been futile while C. Gracchus was still alive. He
managed to intervene in Minucius' assembly and probably no vote was
taken. The senate's impotence is also illustrated by the Lex Licinia de
sodali tatibus. Contrary to common opinion there is no evidence that it
was directed against Clodius' political clubs.
65
We are therefore left with
the telling fact that after the ban on collegia was scrapped in 58
apparently no attempt was ever made to curb Clodius' organisation by
means of legislation; it was probably deemed pointless even to try.
Most suggestive, however, is the strategy used by the senate for the
restoration of Cicero in 57. Clodius had successfully neutralised at-
tempts to pass a bill in the comitia tributa, also using violence and
physical obstruction. The senate's response was exceptional; for the rst
time since the third century the centuriate assembly was called to pass a
legislative bill. The senate even issued a decree urging members of the
Italian elite to turn up on this occasion, which many of them did.
66
The
implication of this strategy has not been fully realised. The senate's
initial failure to pass the bill was not simply due to Clodius' use of force:
that could also have been applied in the comitia centuriata. This assembly
was called because of its different social composition, which reduced the
inuence of Clodius' followers, Thus, even in this exceptional case
where a bill was actively promoted by Pompey and the large majority of
the senate, it was impossible for its backers to stump up sufcient
support among the urban plebs to vote down the followers of Clodius.
In conclusion, it seems that the legislative comitia had never been
directly controlled by the senate. Traditionally they had been frequented
by small groups of people with an outlook and interests largely similar to
those of the political class itself. The negligible attendance of the plebs is
clearly brought out by the tiny scale of the political institutions them-
selves. It is also a question of whether the majority of the plebs was
covered by any comprehensive system of social control and political
exploitation; in a context of minimal popular participation, the elite had
little incentive to create or maintain such a network. This situation had
the potential for dramatic upheavals in the political order; simply by
turning up in the assembly the plebs would be able to take control of the
legislative process. However, the task of bringing representatives of the
plebs into the domain of legislation would have to lie with members of
the elite, which had long monopolised the political initiative. Because
65
See appendix pp. 14951.
66
The alternative would, as Cicero suggests, Att. 3.23.14, have been bribery, the use of
a `comparata multitudo', cf. Sest. 127. In Leg. 3.45 Cicero implies that the comitia
centuriata was called because the law dealt with an individual.
the plebs naturally remained outside the sphere of politics, it was
dependent on members of this class to take the lead.
Therefore, what we are witnessing in the late republic is not so much
the breakdown of social control as the growing disintegration of the
internal cohesion and consensus which had traditionally prevailed
within the elite. For as North has noted: `The popular will of the Roman
people found expression in the context, and only in the context, of
divisions of the oligarchy.'
67
It was this development which in turn led to
the partial mobilisation of the plebs. The process was greatly facilitated
by the inux of rural tribesmen to Rome. But we should be cautious not
to posit any direct causality between the two processes; mobilisation was
not a result of immigration nor did it rely on it in order to succeed. The
potential for `popular' legislation had probably always existed; only the
political will to exploit it had been absent. All that changed, however,
when the senate's authority weakened in the late second century and it
increasingly lost control over its own membership.
67
(1990) 285.
5 Elections
The focal point of Roman politics was the annual elections of new
magistrates.
1
The appointments of quaestors, aediles, praetors, consuls
and censors, in addition to a large number of minor ofcials, were
central events in the public calendar of the Roman republic. And
judging from the scale of the electoral facilities in the Campus Martius
they clearly attracted larger crowds than the legislative assemblies,
which convened in the smaller venues of the Forum and the Capitol.
Likewise the procedures used in the electoral assemblies were less time-
consuming, thus allowing more people to vote in a single session.
Nevertheless, the overall level of participation remained low, and we
may wonder what made a small section of the population take part in
elections, while the large majority stayed away. The aim in this chapter
is to investigate the nature of electoral participation and the different
models which have been used to explain it. Our sources tend to convey
the impression that elections were matters of general interest and
concern among the Roman citizenry during the late republic. The
sources, however, also reect the views and preoccupations of the one
group which was itself directly involved in ofce-holding and the
exercise of power. The question is therefore whether these concerns
1
Millar (1998) 204, 206, 217, has insisted that voting on legislation was the `most
important type of collective decision' and the central focus of Roman politics. It may
seem futile to debate which aspect may have been the most important, especially since
there are no agreed criteria by which to settle such a debate. But it is interesting to note
how this perception is closely linked to the overall approach to the issue. Millar's analysis
is structured as a dynamic narrative, which tells the story of the fall of the Roman
republic and the crucial role played by the people therein. In this particular narrative the
legislative process represents an important factor, whose discursive/rhetorical aspect also
ties in well with the overall emphasis on `democratic' participation. Elections, on the
other hand, did not involve speeches to any signicant extent, and do not offer any
progressive narrative either. From a structural point of view, which looks at Roman
politics as a continuous practice, it is clear that the main activity of the political class
involving broader sections of the populace focused on the annual appointment of new
magistrates. Vast amounts of time, energy and economic resources were invested in this
collective project, which directly or indirectly affected all members of the propertied
classes. Thus also Flaig (1995) 78.
90
Elections 91
were particular to this social group, or, in other words, whether the
appointment of new magistrates attracted much attention outside the
political class itself.
It was suggested above that these important events, to which all
citizens had equal access, were attended by only a comparatively small
number of voters. The capacity of the voting facilities allowed just a
fraction of the citizen population to attend the elections. The Augustan
Saepta could probably accommodate a maximum of about 25,000
voters, which may even have represented an expansion of the original
republican structure. There are, moreover, virtually no reports of over-
crowding having been a problem in Roman elections. A single, excep-
tional case is known from the late republic. In 122 rural voters who had
come to support C. Gracchus could not gain access: instead they seem
to have shouted into the Saepta from the surrounding rooftops, Plut. C.
Grac. 3.1. The story leaves many questions open; their access might
have been obstructed by Gracchus' opponents, or they were in fact not
citizens with full voting rights; obviously they did not try to deliver
their vote by shouting, since the written ballot had already been
introduced. Later, as we saw, there are several indications of very small
numbers attending the elections; for example the brevity of the voting
and counting at the consular election in 45, and the small number of
votes won by Laterensis in the tribus Voltinia in the aedilician elections
of 54.
In a modern democracy such a low level of participation in a general
election would call for an explanation. In a Roman context, however, it
may be more appropriate to turn the question around and ask why
anyone outside the narrow circles of the ruling elite ever took the trouble
of turning up.
The question has rarely been asked, probably because the debate has
followed the agenda found in sources produced by the political class.
Moreover, the preoccupation with elections, expressed in the writings of
the Roman aristocracy, may, somewhat ironically, have been reinforced
by modern experiences and perceptions; the right to vote is now seen as
a coveted privilege and its general extension a sign of democratic
progress. Viewed in that perspective people who enjoy the vote are
naturally expected to make good use of it and exploit any opportunity to
inuence the running of their state. It follows that participation, as
opposed to abstention, does not require any specic explanation. That
logic, however, is a modern one which cannot, for a number of reasons,
be applied to Roman elections.
Firstly, the elected ofcials had a very different role from present-day
politicians, who act as parliamentarians and legislators. Because a
92 Elections
parliament in the modern sense did not exist, Roman politicians had no
legislative powers; that remained the prerogative of the assemblies, in
this period the comitia tributa and the concilium plebis.
2
As magistrates
they formed an executive body some of whom had a right to put new
legislation before the assemblies.
3
Secondly, the scope of politics was fundamentally different in ancient
Rome. Simply because the scale of the state was so much smaller than in
modern societies, politics concerned itself with many fewer aspects of
the lives of ordinary citizens than is the case today. Needless to say,
social and economic policies, as we know them, did not exist in Rome.
Few political issues therefore had implications which reached much
beyond the elite, or had any noticeable effect on the living conditions of
the population as a whole. It follows that the relevance of politics, and
elections in particular, would not have been self-evident to the large
majority of voters.
Political apathy may be overcome by campaigning. In Rome,
however, the canvassing efforts which preceded the elections would
probably have done little to raise political interest among ordinary
voters. They were fundamentally apolitical in their nature and hardly
ever conducted on the basis of specied policies or an articulated
ideological platform.
4
Despite the right of some ofcials to propose
laws, very few legislative programmes or individual bills seem to have
been presented in advance of an election. Differences in opinion
2
On this and the following observations see also Brunt (1988) 356.
3
It is important to note that the magistrates, unlike modern politicians, did not
`represent' the voters to whom they owed their ofce. Roman politicians therefore had
no constituency in the modern sense. Ideally, popular tribunes may have been expected
to protect and further the interests of the plebeians, but they did not represent them, nor
were they directly responsible to them.
4
Cf. Meier (1956) 598, noting that: `. . . in den ro mischen Wahlen die politische
Gegensatze in der Regel keine Bedeutung hatten'; id. (1980) 1123. A view also
expressed by Veyne (1990) 223, who stated that: `what was really at stake in elections
was not anything of importance to the electors'. Cf. Syme (1939) 149; Taylor (1949) 8,
13, 15, 64; Pina Polo (1996) 105; Jehne (1995) 60. A possible exception may have been
the elder Cato's campaign for the censorship in 184, Livy 39.41.14; Plut. Cato Maior
16. Millar (1998) 35, 191, has argued that electoral campaigning was often conducted
on specic policies. The supporting evidence is very limited, however. Clodius, in case
he really did announce policies in advance of the praetorian election of 52, is likely to
have been an exception. Spielvogel (1997) argues that Clodius had revealed none of his
`popular' legislative programme in advance of his tribunician election. Pompey, as Millar
notes (1998) 64, was already designate when he declared his intention to restore the
tribunician powers. Similarly, Rullus only drew up his reform plans after his election,
Cic. Agr. 2.12, cf. Millar, ibid. 103. Drummond (1999) has shown that the `tribunician
programme of 63', suggested by Dio 37.25.34, is a mirage. Dio 40.61.2 does not
suggest that Curio had publicised his proposals before taking up his tribuneship.
Yakobson (1999) has recently taken Millar's argument even further, but the evidence
remains slight.
Elections 93
between candidates were not emphasised in the electoral contests either;
to the extent that they existed they were generally not spelt out during
the campaigns, in which only personal vices and allegations of bribery
were used as political weapons, cf. Comm. pet. 52. `Politics', it seems,
was plainly avoided in the run-up to elections.
5
As Cicero was expressly
advised in the Commentariolum: `Yet, during your canvass, you must not
deal with politics either in the senate or in the contio.'
6
The political
conicts of the late republic, often provoked by `popular' tribunes,
occasionally revolved around issues of interest to broader sections of the
electorate. But even in these cases there are few signs of the issues
having been raised in an electoral context. In general, `popular' policies
were presented only after the assumption of ofce. Campaigns for the
tribunate were therefore not fought on what we would consider political
issues either. The àpolitical' character of elections is further suggested
by the fact that the lower, largely administrative posts, e.g. quaestors
and the vigintisexviri, which played practically no political role, were
lled by the same `democratic' procedure as the powerful, higher
magistracies with the right to propose legislation. Altogether this serves
to underline the important fact that Roman elections did not present the
voters with a range of political options; issues that might cause dissent
were generally banned from most electoral contests, as were specic
pieces of legislation and even appeals to particular interest groups. The
choice, it seems, was entirely one between different individuals, all
drawn from the same social class.
Again a comparison with classical Athens may be helpful in putting
the peculiarities of Roman elections into perspective. Thus, while most
Athenian magistrates were chosen by drawing lots, in principle the most
democratic method, other posts which required special skills were lled
by a vote in the assembly.
7
These elections differed on crucial points
from the Roman system. Most importantly it was open to any Athenian
to put himself forward as a candidate, citizenship being the only
5
Political positions were indicated only in relation to other leading politicians, e.g.
Pompey, not on issues, cf. Comm. pet. 14; 51.
6
Comm. pet. 53: `nec tamen in petendo res publica capessenda est neque in senatu neque
in contione'. In an attempt to demonstrate the profoundly political nature of elections,
Yakobson (1999) 1525, has tried to reinterpret this unequivocal piece of advice,
arguing that it was directly linked to Cicero's vulnerable position as a homo novus, which
forced him to tread more carefully. However, if elections were run and decided on
political issues, it is difcult to see how it would improve Cicero's chances to remain
outside this debate and appear apolitical. It is not clear why a new man should be
particularly cautious in fact the very concept of novitas becomes meaningless if ofcials
were elected on the basis of policies rather than family prestige and personal
connections.
7
Hansen (1987) 1203.
94 Elections
qualication required. Moreover, to ensure that the formal right to hold
ofce was also a practical possibility, state ofcials received remunera-
tion, thereby allowing men without property to stand. The contrast with
Rome is striking. Here the ofcial property qualications meant that all
magistrates belonged to the same social stratum of the very rich, in
effect reducing the elections to a choice between one noble and another.
These limits to the popular choice were further accentuated by the
practicalities of Roman voting. Firstly, the elective process was very
time-consuming, and unlike Athens Rome offered no remuneration
to plebeian voters. People could expect to spend a whole day at the
Saepta. Occasionally the assembly even had to reconvene on the
following day to complete the vote.
8
That happened when the statutory
number of magistrates had not reached an absolute majority of the
electoral units. The consular election of 45, as we have seen, lasted less
than four hours, but that was clearly exceptional. On this occasion the
assembly voted for only one candidate, nominated by Caesar; it was
therefore already over when the rst two classes had delivered their vote.
The crowd had, moreover, originally convened as a comitia tributa,
expecting to elect the quaestors, which may have affected its size and
social composition.
Secondly, the workings of the assemblies contrasted sharply with the
Athenian ecclesia, where all votes had equal weight. In Rome the system
of corporate voting allowed the inuence of individual voters to be
carefully graduated. As we have seen, urban plebeians, and freedmen in
general, were inscribed in only four tribus. The comitia centuriata was
even more heavily tilted in favour of the propertied classes. While the
senators, the knights and the rst class held 88 out of 193 centuriae, the
two lowest classes were squeezed into 30 centuriae and the proletarii into
just a single one. Moreover, since the freedmen did not serve in the
legions, their representation in the comitia centuriata, formally the
Roman citizenry organised as an army, was minimal. They were prob-
ably all inscribed in the ve non-armed centuriae.
9
8
Examples include Livy 22.35; 37.47.7; 40.59.45. In Mur. 35 Cicero refers in a more
generalised way to `dies intermissus aut nox interposita'.
9
Apart from the proletarii, the non-armed centuriae also included two centuriae of artisans
and two of musicians, trumpeters and horn players, cf. Taylor (1966a) 86, 155 n.38;
Ku hnert (1991) 21. Against freedmen's allocation to these units, Treggiari (1969)
1667 refers to Comm. pet. 29, where Cicero is advised to seek urban support from:
`Multi homines urbani industrii, multi libertini in foro gratiosi navique versantur.' This
is the only mention of freedmen in the context of consular elections in the
Commentariolum and elsewhere and it probably refers to their usefulness in raising
general support through their gratia and daily activity in the Forum. Cf. the anecdote
about Scipio Aemilianus, Plut. Aem. Paull. 38.3, in which freed supporters feature, not
as inuential voters, but perhaps anachronistically as crowd xers.
Elections 95
This hierarchy of voting not only meant severely restricted inuence
for some groups; the strict order of voting in the comitia centuriata also
implied that the lower classes might not get to vote at all. Yakobson has
recently tried to upgrade the inuence of the plebs in this assembly,
arguing that whenever there was a split in the vote of the elite, the lower
centuriae had a real say.
10
He also points to the repeated sessions as cases
where even the proletarii would have come to vote. That, however, was
not a decisive vote, and for the voting to reach the proletarii a very
profound split in the elite and the higher classes was required. Clearly
these examples are not typical; Cicero explicitly noted that the lower
classes had no certainty of being called to vote, Mur. 71. Cicero's own
election to the consulship illustrates the limits to their inuence;
supported by all the propertied centuriae, he was elected when just
ninety-seven centuriae had declared their vote, Off. 2.59. Incidentally,
the voting continued until most centuriae had been called, though it
nished before reaching the proletarii. But such a deep split in the vote of
the elite was entirely a matter of chance: low-ranking citizens could
never predict whether it would be worthwhile for them to turn up.
Thirdly, the geographical centralisation of the political process had
obvious implications for people's ability to take part. More than any-
thing this aspect serves to remind us just how far from modern
democratic ideals were the realities and underlying rationality of the
Roman assembly. The simple fact that the citizens of a sizeable terri-
torial state could deliver their vote only in one topographical location
illustrates the extent to which large-scale participation was deliberately
precluded through the institutional framework. Also prior to the Social
War many Romans had had to make long journeys in order to exercise
their voting rights, for example the Roman colonists settled in Cisalpine
Gaul. But with the enfranchisement of the Italian allies, which was
implemented in 70, the situation plainly became absurd. By then the
large majority of the citizens lived too far from the voting facilities ever
to be able to participate. And since the elections of different magistrates
were held at separate sessions, extra-urban voters had to travel to Rome
not just once but several times over if they wanted to inuence the
electoral assemblies.
In sum, mass participation was discouraged by a number of factors.
Quite fundamentally, the àpolitical' nature of the elections would
probably have made the whole exercise an irrelevance to the large
majority of the population. Its alienation from the electoral process
would have been further reinforced by the practicalities of delivering the
10
Yakobson (1992).
96 Elections
vote: the laborious and time-consuming procedures and the lack of
economic compensation offered, the graduation of votes and electoral
inuence, and the extreme centralisation of the voting procedure. These
factors may easily explain the low level of attendance we nd in Roman
assemblies. Still, some voters did turn up. Their social and regional
background cannot be specied, and although the elite obviously would
have turned out in much greater strength, the plebs must to some extent
also have been represented. The question is therefore why these voters,
unlike most of their fellow citizens, chose to make this sacrice.
Traditionally their participation has been explained by reference to
the clientela model.
11
Although primarily designed to explain not so
much why people turned up as how they behaved when voting in the
assembly, this model does provides an answer to the former question.
The clientela model puts overall emphasis on the ability of the elite to
control the popular vote by the use of individual patronal powers. On
this interpretation there was little room for personal choice; involvement
in politics was neither voluntary nor independent. People turned up out
of obligation in order to vote for their patron or a candidate who enjoyed
his support. Thus, our picture of popular participation and voting
becomes dominated by personal ties and obligations, which in effect
reduced elections to a contest between noble factions and their armies
of personal dependants. In this way the theory explains not only how the
elite were able to control the electoral assemblies with such apparent
ease but also why people were there in the rst place.
As already noted in the previous chapters, important objections have
now been raised against this model, whose evidential foundations seem
to crumble when subjected to more detailed scrutiny. The harder one
looks at the evidence the more difcult it has proved to nd traces of this
vertical division of society in the politics of the Roman republic. As a
social category clients appear conspicuously absent from the political
scene.
12
In the entire Commentariolum petitionis clients are mentioned
only once; in a routine listing of relevant voter groups they feature after
tribules and vicini, 17. This important text does not contain any allusions
to tied voters either, and several passages clearly imply that clients were
not considered a decisive factor. That is also the conclusion offered by
Cicero's discussion of the candidates for the consular election in 54, Att.
11
See above pp. 6778.
12
A rare statement as to the importance of clientela is found in Sall. Iug. 85.4: Àd hoc, alii
si deliquere, vetus nobilitas, maiorum fortia facta, cognatorum et adnium opes,
multae clientelae, omnia haec praesidio adsunt.' The passage appears in a ctitious
speech by Marius. As such it is highly tendentious and likely to exaggerate the aspect of
inherited privilege. Moreover, the nature and social composition of these clientelae are
not dened.
Elections 97
4.16.6. In his survey of their relative strengths and prospects Cicero
never raises the issue of clientela. `Friends' are noted in the case of
Domitius, but that is hardly a reference to humble clients. In the same
breath the electoral benets Domitius might derive from his games are
mentioned, again questioning the importance of personally tied voters.
Other factors considered by Cicero are the candidates' wealth, previous
municence, paternal popularity, support from Caesar's troops and the
endorsement of Pompey. Thus, Cicero gives no hint whatever that the
consular election of 54 would have been a simple contest between
alliances of the elite with their personal dependants as mere voting
fodder.
Two aspects of the elections themselves would seem to cast further
doubt on the idea of clientela as a decisive factor in the electoral process:
rstly, the methods and importance of campaigning, and secondly the
patterns of voting observed in the elective assemblies.
The extent of campaigning in the run-up to elections and the serious-
ness with which it was conducted in the late republic square badly with a
model of universally tied voters.
13
Electoral campaigning was a central
feature of public life in the late republic, which in its scope and
implication went far beyond a mere rounding up of personal dependants
and allies. Canvassing took place at all levels and employed a wide range
of tactics and devices. Voters were targeted individually, approached
collectively, often through intermediaries, and broader sections of the
population were courted by costly municence, which became a vital
component in electoral success.
14
Games were given at huge expense,
public dinners held, and donations made to the tribes, leaving many
politicians heavily in debt.
15
Candidates also showed great concern with
their public image, which was perceived as a potential electoral asset
or liability; it was clearly important to be known as a generous and
respectable character. This broad range of activities would seem to
make sense only on the assumption that voters were sufciently inde-
pendent to be susceptible to personal or collective persuasion. That is
also the implication of the widespread bribery which became a charac-
teristic feature of late republican elections. Altogether this would
suggest that cliental ties were not nearly as universal as has previously
been assumed; or in any case not strong enough to prevent clients from
13
Cf. Millar (1984) 1014.
14
E.g. Cic. Fam. 2.6.3, which lists the support won for Milo's consular candidature in 53:
`Habemus haec omnia, bonorum studium . . . vulgi ac multitudinis propter magni-
centiam munerum liberalitatemque naturae . . .'
15
Most famously Caesar incurred huge debts during his aedileship, cf. e.g. Plut. Caes.
5.3.
98 Elections
being otherwise persuaded or votes from becoming in Brunt's expres-
sion a `marketable commodity'.
16
Extensive campaigning was perceived as essential to electoral success.
That is the obvious rationale behind the advice offered in the Commen-
tariolum. And Cicero himself, in Pro Plancio 9, explained Laterensis'
defeat by his reluctance to canvass, declaring that the populus èlects
those who court it most assiduously'.
17
The younger Cato may also have
lost his bid for the consulship because of his refusal to campaign, Plut.
Cato Min. 49.4.
A striking feature of late republican elections is the genuine uncer-
tainty about the outcome of these events. The sources frequently refer
to the unpredictable, even irrational nature of the electoral assemblies in
this period. In Pro Murena Cicero exclaims that: `Nothing is more ckle
than the masses, nothing harder to discover than how men intend to
vote, nothing trickier than the whole way in which elections work.'
18
This and similar statements, for example Planc. 12, 512; Mur. 35; 53,
may be linked specically to the argument Cicero was pursuing in these
ambitus trials; in his attempt to explain the defeats of the prosecutors,
Cicero had an obvious interest in stressing the erratic aspect of elective
assemblies. Still, his assertions were probably not widely off the mark,
intended as they were to inuence a panel of judges familiar with
Roman politics. And the same uncertainty also surfaces in his private
correspondence, for example Q. Fr. 3.1.16, from 54, where he states
that: `So far it is extremely uncertain both when the elections will be
held and who will be elected.'
19
Certainly, the preferences of the elite did not always prevail. Noble
grandees could lose to less prominent opponents, and there are exam-
ples which clearly demonstrate that the nobility was powerless to
prevent the victory of a candidate who had gained wide popularity.
Thus, in 148 Scipio Aemilianus was elected to the consulship before the
prescribed minimum age against the expressed wishes of the nobility.
20
Marius too overcame noble opposition to his consular candidature in
108.
21
Later, in 105, he even repeated the feat and was elected in
absentia within a decade of his rst consulship, Plut. Mar. 11.1. Cicero
16
Brunt (1988) 127.
17
`. . . facit eos, a quibus est maxime ambitus'.
18
Mur. 36: `Nihil est incertius volgo, nihil obscurius voluntate hominum, nihil fallacius
ratione tota comitiorum.'
19
Àdhuc erat valde incertum et quando comitia, et qui consules futuri essent.'
20
Livy per. 50; App. Pun. 53033, cf. Astin (1967) 5960; Develin (1978). Also the
praetorian election of 184 illustrates the electorate's ability to defy the senate, Livy
39.39.
21
Sall. Jug. 73.7: Ìta perculsa nobilitate post multas tempestates novo homini consulatus
mandatur.'
Elections 99
also lists several cases where quite unknown or far less prominent
candidates defeated members of the nobility in consular elections, and
gives examples of consuls who, at an earlier stage in their careers, had
been passed over for lower ofces.
22
The tight control over the electo-
rate, assumed by the clientela model, is called into question by these
instances. There are few signs in this period that elections were deter-
mined by the patronal powers of old senatorial families, or xed in
advance by internal horse-trading within the elite.
Other aspects of elections may suggest that the voters were not just
independent-minded but downright volatile in their behaviour. The
centuria which happened to open the poll was known to exercise a
disproportionately strong inuence on the outcome of the election. The
praerogativa centuria was drawn by lots from amongst the centuriae of the
rst class juniores and voted separately before the rest.
23
The importance
of its vote was widely recognised in antiquity. Thus Cicero claimed,
probably exaggerating, that: `the century which votes rst carries of
itself such weight that no candidate for the consulship has ever secured
its vote without being ultimately declared rst consul either at that very
election or at any rate for the following year'.
24
The signicance of
winning here is further underlined by the bribery case from 54, when
Memmius and Domitius Calvinus promised no less than 10 million
sesterces to the praerogativa centuria, Q. Fr. 2.15b.4.
25
The vote of the
rst centuria was seen as an omen,
26
but the phenomenon may also be
explained as a bandwagon effect, whereby people, eager to vote for a
winner, tended to follow the example set by the inuential rst unit.
27
Still, whichever interpretation is favoured this pattern suggests a remark-
able volatility among the voters, who appear to have cast their vote
according to the chance outcome of the rst centuria. This willingness of
the crowd to follow any incidental lead that was offered seems a strong
argument against the notion of the electorate as a mere puppet in the
hands of noble patrons.
28
Finally the case might be argued on the basis of numbers. For how do
22
Planc. 12; 512; Mur. 36. Cf. the section in Val. Max. 7.5, on electoral defeats: `de
repulsis'.
23
Meier (1956).
24
Planc. 49: ùna centuria praerogativa tantum habet auctoritatis, ut nemo umquam prior
eam tulerit, quin renuntiatus sit aut iis ipsis comitiis consul aut certe in illum annum';
cf. Livy 26.22.13: àuctoritatem praerogativae omnes centuriae secutae sunt', Meier
(1956) 593.
25
Cf. Cic. Verr. 1.26; Schol. Gronov. 350 (St).
26
Cic. Mur. 38; Div. 1.103.
27
Meier (1956) 593; Taylor (1966a) 76.
28
Meier (1956) 5978 noted the difculty of reconciling the importance of the
praerogativa centuria with elite control over individual electoral units.
100 Elections
we explain that so few voters/clients turned up on occasions of such
obvious importance to their patrons? Elections have traditionally been
seen as the primary occasions on which clients could pay back favours
received.
29
The notion of clientela as a reciprocal system of benecia and
ofcia naturally puts focus on the popular vote which stands as the
clients' foremost resource in this mutual exchange. A decline in clientela
is commonly posited after the fall of the republic when elective comitia
became a formality and the votes of the plebs lost their value to the elite.
However, it remains a fact that even at the very peak of elite competition
in the late republic only a fraction of the citizenry ever attended the
elections. That aspect has yet to be satisfactorily accommodated into
those models which operate with individual voter control as a determi-
nant factor in Roman elections.
In sum, it seems that the politically active citizens were not only too
few, but also too volatile and open to persuasion; in short, too indepen-
dent to t into a conventional model of patronal control. This situation
might on principle be linked to the supposed decline of political clientela
in the late republic when the system is assumed to have been in a state of
advanced disintegration, perhaps exacerbated by the introduction of the
secret ballot which offered a cover against direct patronal control. But
even on this explanation it remains an uncomfortable fact that the only
period of the Roman republic for which we have more detailed evidence
does not seem to t the model. At the very moment when the sources
allow us a more nuanced picture of participation, campaigning and
voting behaviour, a situation emerges which is far too complex to be
explained by a simple clientela model.
We are left with the conclusion that political participation in this
period will have to be interpreted along different lines. The question is
where we go from here. One option is to abandon completely the
concept of patronage as a means of understanding Roman elections and
instead interpret them as democratic processes in which people chose
more or less freely between candidates according to their personal
preferences. That model is based on the assumption that if voters were
able to act independently of nobles and politicians, themselves reduced
to courting popular favour, the process is automatically endowed with
`democratic' qualities; elections become a viable means of expressing
genuine popular views and options. Popular participation on this view
becomes the result of individual voters personally engaging in the
political process.
30
29
Thus, Wallace-Hadrill (1989) 81 implies that during the republic the main pay-off for
the patron was the client's vote.
30
This is broadly the view taken by the `democratic' school, cf. ch. 1.
Elections 101
However, the logic of this inference from independent voters to a
democratic process hangs on the questionable notion of magisterial
elections being inherently relevant to the mass of Roman citizens. But,
as argued above, it is a big question whether elections were `political'
events in the sense that ordinary citizens would have perceived any
personal interest to be at stake in the outcome. Most plebeians could
have been rightly excused for seeing their participation as a pointless
waste of time and effort. Our opening question is therefore left un-
answered, and we still have to explain why members of the lower classes,
however independent, would turn up for long exhausting sessions to
make a largely irrelevant choice between different members of the
elite, without even having any certainty of getting to deliver their vote.
The pattern of voting, and in particular the importance of the
praerogativa centuria, throws further doubt on the democratic model.
For again we may wonder why a small, supposedly more politically alert
section of the plebs would turn up if their convictions were so weak that
the omen of the rst centuria was habitually followed. The notion of
political engagement as the driving force behind their participation
makes little sense if the active citizens decided how to vote only when
they were already waiting inside the Saepta.
In sum, the independence of the voters does not in itself turn the
elections into a `democratic' process; there have to be issues involved of
some basic interest and importance to the electorate.
31
If the outcome
was largely irrelevant to the lives of most citizens, it seems futile to stress
their independent choice. Popular participation, to the extent that it
occurred, will therefore have to be explained by factors other than
engagement in political issues; in general we have no reason to believe
that the plebs had particular preferences in the elections. There are
obvious exceptions to this rule. The popularity of victorious army
generals like Scipio, Marius and, to some extent, Pompey is evident, and
charismatic popular leaders like Clodius also managed to establish a
personal power base among the urban plebs. Still, they remain excep-
tions, and their popularity was often eeting and hard to sustain over
longer periods, as C. Gracchus' failure to secure a second re-election in
121 well illustrates.
Any attempt to explain popular participation must in other words
begin with the recognition that plebeians had no natural part in these
events, at which their attendance was discouraged by a number of
institutional and practical factors. Their patchy turnout is therefore
31
By insisting that: `. . . the Roman voter . . . could thus feel free to exercise his suffrage
freely', Yakobson (1995) 136 misses the point that no matter how free the vote was, the
overwhelming majority of the citizens never bothered to turn up to deliver it.
102 Elections
most plausibly interpreted as a result of the lack of political incentives
for the plebs to overcome these obstacles. On the other hand, it follows
that those who did turn up must have had specic reasons for doing so.
And since elections generally had little political substance, these reasons
would have to be personal in nature. Logically their appearance would
have been promoted by members of the political class, for whom the
elections had a real urgency. For while the plebs had the formal powers
to appoint the leaders of the state, it had little interest in exercising
them. Popular participation would therefore seem to be a question of
the latter persuading the former to turn up.
Putting overall emphasis on the aspect of elite mobilisation, this
model of electoral participation ties in with the interpretation of the
contiones and legislative assemblies suggested in the chapters above. Like
the clientela theory it shifts the initiative away from the voters and onto
the candidates; that in turn leaves it open to some of the same objections
that were raised against this model.
For while the general turnout would suggest that the few active
voters, unlike the large majority, were strongly motivated, they also
appear to have cast their votes on sudden impulses and somewhat imsy
grounds. The importance of the vote of the rst centuria has already
been mentioned. And it was also an acknowledged fact that a prestigious
family background represented a valuable electoral asset; the sheer
grandeur of the name was supposed to attract the votes. Still, we might
wonder whether people would really take the trouble to turn up simply
because one of the candidates came from a famous family. Again we are
faced with the problem of reconciling a general model of elite mobilisa-
tion with the apparent volatility and superciality of the voters. The
electoral campaigning contains similar paradoxes. Here we nd both
personal canvassing, targeting individuals in the Forum and among the
social circles of the elite, and much more generalised campaigning
which simply aimed at bringing a positive image across to the public.
For it would be surprising if a supercial impression of a candidate's
character, however positive, was sufcient to break the indifference of
the electorate and make them turn up for the polls. We are faced with an
apparent contradiction between the low level of participation and the
seemingly frivolous way in which votes were cast.
Part of an answer to this problem may lie in an evaluation of the basic
technicalities of voting. One issue seems crucial: the number of votes
each voter had at his disposal. In principle the voter could either put
forward as many names as there were posts to be lled, or he could
support just a single candidate in each election. There has been little
debate on this question and in general the former has been assumed to
Elections 103
be true. Hall was the rst to raise the alternative possibility, but did not
commit herself to it.
32
To my knowledge only Nicolet has opted for this
solution.
33
Although these two options would seem to exhaust the
possibilities, I shall argue that in practice a compromise is also viable,
indeed persuasive.
In support of the one-vote model Nicolet has referred to Varro's De re
rustica, which is set during an aedilician election, while people were
waiting for the result to be announced by the Villa Publica next to the
Saepta. The dialogue makes repeated reference to òne's candidate'.
First a participant mentions that they `wished to be on hand to escort
the candidate whom we were supporting, when he returned home', and
later, when accusations of fraud were made, `Pavo arose, as it was the
attendant for his candidate who was reported to have been arrested.'
34
Thus, Varro suggests that his characters were associated with only one
candidate; presumably they had delivered just a single vote in his favour.
Nicolet's is a very important observation, but as we shall see, it may not
necessarily indicate that voters could vote for just one candidate. In
addition Nicolet notes that the occasional failure to elect a whole
magisterial college in a single session is best explained if people only had
one vote.
35
Supposedly some centuriae or tribus would in that case not
always have been able to nominate sufcient candidates for all the posts.
The arguments against this model are cogent, however. Most
unequivocal is the practice of coitio, whereby two candidates joined
forces in order to share their supporters and mutually benet from each
other's mobilisation of voters.
36
The practice was not in itself illegal, but
was considered unfair to the other candidates whom the arrangements
were designed to keep out. A well-documented example comes from the
aedilican election of 54, when Plancius was accused of having entered
into an alliance with Plotius. Together they had allegedly bribed the
voters from the tribes of Plancius' native region in central Italy. The
argument seems to have revolved around the fact that the two had
received an almost identical number of votes in these tribes. The logical
32
Hall (1964) 297. In (1998) 27, Hall hesitantly rejects the idea.
33
Nicolet (1970) 12930; (1980) 274.
34
3.2.1: èt candidato, cui studebamus, vellemus esse praesto cum domum rediret', and
3.5.18: `Pavo surgit, quod eius candidati custos dicebatur deprensus.'
35
Nicolet (1980) 274. The known instances of repeated voting sessions are listed in Hall
(1972) 11 nn.1718.
36
Cic. Att. 1.17.11; 4.15.7; Q. Fr. 2.15b.4; 3.1.16; Cluent. 148; Planc. 534; Parad. 46
(involving common bribery), Asc. 83. Cicero may have entered into some kind of
electoral pact with Antonius when they ran for the praetorship. Cicero claimed that
Antonius came third due to `concessione competitorum et collatione centuriarum et
meo maxime benecio', Cic. ap. Asc. 85C.
104 Elections
inference is that Plancius' supporters had voted for two candidates, not
one. Alternative explanations, envisaging a division of the supporters
into two separate groups, instructed to vote for different candidates,
become excessively complicated. And even if it might have been possible
in the case of Plancius' local supporters, it is inconceivable that coitiones
could generally have worked along these lines. It is difcult to believe,
for example, that in 64, when Catiline and Antonius joined forces, Asc.
83C, the former asked some of his supporters to cast their votes not for
himself but for Antonius, merely on the expectation that Antonius
would do the same for him. The more obvious explanation of coitio is
plainly the multiple vote. It is also worth noting that later evidence from
Pompeii clearly suggests that people voted for as many candidates as
there were posts.
37
On this fundamental point municipal voting prac-
tices are unlikely to have differed radically from urban traditions; and a
later change under the principate is hardly plausible.
Nicolet's argument from the repeated voting sessions carries little
conviction. In principle a single vote would sufce to put a candidate's
name on the list of nominees. It therefore seems unlikely that some units
even under a single-vote system would not have been able to
nominate the required number of candidates. At consular elections,
where only two posts were to be lled, a failure to do so would
presuppose the complete unanimity of a centuria.
38
When it came to the
larger magisterial colleges, the quaestors in particular, problems might
occur. But here the electoral units, that is the tribes, were also much
larger, making it relatively easier for candidates to raise support in each
unit. Even quaestorian candidates should have been able to mobilise a
single voter in each tribus, which was all it took to get their name on the
list; and usually we hear of repeated sessions only for the higher ofces.
The failure to conclude the vote in a single assembly was therefore not,
as held by Nicolet, caused by insufcient numbers of votes cast in each
unit, leading to a shortage of candidates nominated. The cause of the
problem was the practice of electing magistrates by an absolute instead
of a relative majority. Under this system a deep split in the vote would
automatically prevent the required number of candidates from getting
elected in the rst round.
Further evidence for the multiple vote comes from Comm. pet. 12,
where Q. Cicero assured his brother that his electoral prospects were
37
On elections in Pompeii see Mouritsen (1988).
38
Thus, Nicolet, (1980) 274, is mistaken in asserting that the failure to elect a full college
`could not happen on any other hypothesis' than the single vote, adding that ìf each
citizen had had to vote for two candidates it is hard to imagine that a second
nomination would not have emerged . . .'
Elections 105
very promising indeed.
39
Having stressed the unsuitability of his compe-
titors and listed their incompetence and vices, Quintus asked rhetori-
cally: `quis enim reperiri potest tam improbus civis qui velit uno
suffragio duas in rem publicam sicas destringere?' The voter's ability in
ùno suffragio' to strike `duas sicas' against the public interest must
clearly eliminate any remaining doubt about this question.
40
Quintus'
point was that Cicero could expect support from most of the voters,
since it seemed inconceivable that anyone would cast both his votes for
Cicero's unworthy opponents.
Although the conclusion must be that voters had as many votes as
there were posts to be lled, it does not necessarily follow that all voters
nominated a whole college of magistrates on their ballots. The case of
the quaestors illustrates the point. After the Sullan reform there were no
fewer than twenty quaestors. Apart from the practical difculties in
tting so many names, even when abbreviated, into the ballots,
41
the
time it would have taken makes this scenario quite unrealistic. Even a
cautious calculation suggests that the procedure would have been far
too time-consuming.
42
It seems evident that voters could not have been
expected to ll in a complete list of quaestors; more likely people voted
for just a few candidates. Obviously that did not invalidate their votes.
Since the electoral system did not rely on full lists being submitted, it
was perfectly possible simply to add in the nal count as many names as
each voter had chosen to put down on his ballot.
A exible system of this type would explain how people could dispose
of several votes, while at the same time the participants in Varro's
dialogue seem to support just one candidate. The assumption that a
system of multiple votes automatically implied that all voters cast a full
number of votes is formalistic and practically unfeasible perhaps
39
The Commentariolum is here accepted as an authentic essay written by Cicero's brother
Quintus before the consular election in 64, cf. e.g. Nardo (1970) 12937; Richardson
(1971); Palmer (1971) 38593; David et al. (1973); Jehne (1995) 58 n.41; Fraschetti
(1990) 197; Pani (1997) 139.
40
A very similar phrase occurs in Cic. In toga cand. ap. Asc. 93C: `duas uno tempore
conantur in rem publicam sicas destringere'. As suggested by Richardson, (1971) 441,
Cicero may here have taken inspiration from his brother's essay, which is the more
specic in its reference to the double vote. Rather than being an argument against the
authenticity of the Commentariolum, this passage may therefore speak in its favour.
41
In the Lex repetundarum, 51, the length of a ballot used by the juries is prescribed to be
four digits = 7.5cm. Hall (1998) 29 argues that electoral ballots must have been
considerably larger.
42
Simply to write twenty sets of initials would have taken more than a minute for most
voters. Assuming therefore that the delivery of each vote took about eighty seconds, the
vote of a large tribus of say 500 voters would have lasted more than eleven hours. The
counting would also have been both complicated and time-consuming no fewer than
10,000 votes had been cast by a tribus of that size. The election could not possibly
therefore have been concluded in one day.
106 Elections
inuenced by modern perceptions of the vote as a democratic privilege
which is there to be used to its full potential. However, the ability of
voters to cast more than one vote, often many more, has wide impli-
cations for our understanding of electoral behaviour and voter mobilisa-
tion. The multiple vote may explain some of the curious features of the
elections which were noted above.
43
The level of attendance in elections and their lack of political sub-
stance would suggest that the plebeians who did take part did so for a
specic reason, that is to vote for a particular candidate, to whom they
had promised their support.
44
Such commitments, however, did not
exclude eventual support for other candidates too.
45
Since voters had at
their disposal at least one more vote, often many more, there was also
room for an element of spontaneity. Thus, the secondary, uncommitted
votes may explain how the praerogativa centuria was able to inuence the
voting of all the other centuriae. Voters could decide to cast an additional
vote for a candidate, singled out by this omen, and still remain loyal to
the candidate who was mainly responsible for their attendance. If
second votes were often decided on the spot, that might also help to
explain the importance of some of the other `supercial' factors in
Roman elections. Thus voters might choose to pay tribute to a presti-
gious name which featured in the list of candidates, or award a candidate
who had made a favourable impression during his campaign. This
model of voting thus combines specic motivation for turning up with
considerable scope for free choice and last-minute impulses, when
people were there.
In the nal count, however, all votes had equal weight: those seriously
considered in advance as well as the `secondary' ones which might or
might not be cast, often determined by a last-moment impulse. This
situation may have been responsible for the somewhat heterogeneous
form taken by the electoral campaigns which could narrowly target
43
Another factor affecting the voting of the comitia centuriata relates to the practice of
electing by simple majority of the centuriae, which meant that when a candidate had
reached the required number of centuriae he was taken off the list and the following
classes could no longer vote for him. Those voters who had turned up to support a
particular, successful candidate would then have to cast their vote for someone else,
perhaps chosen almost at random from among the remaining candidates.
44
Brunt (1988) 29 speaks of uncommitted voters. That seems almost an oxymoron in a
Roman context; citizens entirely without commitments were unlikely to turn up at all.
Their voting rights would probably never have been realised. The same applies to
Jehne's `schwankende' voters, (1995) 63, who presumably would not have voted either.
45
Political etiquette might in some quarters have entailed that voters committed
themselves to only one candidate. Thus, Cic. Att. 1.1.1 mentions people who, when
approached by Galba, excused themselves, referring to their obligations to Cicero:
`Nam illi ita negant vulgo ut mihi se debere dicant.'
Elections 107
individuals and aim broadly at the public in general. In the following,
the campaigns will be interpreted in the light of this theory of `primary'
and `secondary' votes, which though equally important had to be
courted by different means and strategies.
Elective assemblies were, unlike the legislative comitia, annual events
for which prospective candidates could prepare long in advance. This
opportunity to conduct carefully planned, extensive campaigns is a
distinctive feature of late republican elections. The importance ascribed
to canvassing is illustrated by Cicero's campaign for the consulship,
which he started planning more than one year before polling day, Att.
1.1.2. As we have seen, electioneering could take a variety of forms. A
primary task was the mobilisation of people personally committed to
your cause. By face-to-face canvassing personal contacts were estab-
lished or reafrmed.
Part of this networking took place in a public or semi-public environ-
ment. A prime opportunity to attract support was the daily salutationes,
where friends and followers were received at the home of the candi-
date.
46
As the day proceeded this ceremony transformed itself into
another social ritual when the candidate, followed by a large crowd of
assectatores, made his way to the Forum. Here the point was to greet as
many people as possible and establish some kind of personal rapport.
The employment of nomenclatores was important to this purpose. They
were assistants specialising in knowing people by name, discreetly
passing the information on to the candidate who could then pretend to
have remembered it himself.
47
The use of nomenclatores was banned at
some point by the Lex Fabia, Plut. Cato Min. 8.2.
48
The practical
usefulness of nomenclatores is curious considering the size of the Roman
electorate. For however brilliant a knowledge of people the nomenclatores
may have developed, they could not possibly have covered more than a
small section of the urban population, let alone of the entire citizenry. It
seems evident that those citizens who were approached in this direct
manner all enjoyed a certain social standing and were inuential in
particular social circles.
The Commentariolum, 16, draws a distinction between the pursuit of
àmicorum studia' and of `popularem voluntate', but the methods
recommended to achieve the latter do not appear to have aimed very
broadly. Even the `populus' was approached personally in the Forum,
received in private, and promised individual favours, 49. The people
who could be found in the Forum, as I have argued above, may not have
46
Cic. Mur. 44; Comm. pet. 345; 37.
47
Cic. Mur. 77; Comm. pet. 28; 412.
48
Bernert (1936) 81720.
108 Elections
been ordinary workers and shopkeepers; more likely they belonged to a
different social stratum. We have no indication that Cicero and other
candidates ever had personal contact with tabernarii or craftsmen. It may
be easy to read modern practices into Roman campaigning. But indis-
criminate `pressing the esh' does not seem to have been part of the
exercise; for all we know candidates did not make a point of publicly
approaching people from all social levels.
Behind the public aspect of personal canvassing the candidates
worked hectically to mobilise friends and acquaintances in Rome, and
perhaps also their contacts in the municipia, persuading them all to
further their cause within their respective circles. These negotiations
behind the scenes revolved around the exchange of favours, past and
future. People were moved àd studium navandum' by `spe ofciorum'
and `recentibus beneciis', Comm. pet. 19. Some were clearly under a
heavy moral obligation to support the candidate; in Cicero's case that
applied particularly to those he had defended in court.
49
But in general
the situation was more complicated; a subtle give and take in which
future support was promised to those with ambitions of their own, while
past beneciaries were reminded of their debt. The overall picture of
electioneering emerging from our sources would seem to conrm the
view of patronage as an essentially uid and unstable relationship, which
was often established or revived for specic ad hoc purposes. Traditional
clientela seems conspicuously absent from electoral campaigns. Instead
we nd an ongoing renegotiation of positions within the elite and a
complex network of short-lived alliances.
Even more important in the context of popular participation is the
fact that the support provided by other nobles was not measured in
clients, but in their personal prestige and willingness to campaign on
behalf of the candidate. Thus, in the Commentariolum the backing by
`homines inlustres honore ac nomine' is highly regarded, since they
`bring a candidate some prestige, even if they do not take an active
interest in canvassing'.
50
That is, despite the importance of active
campaigning, even their passive support would lend dignity and cred-
ibility to the candidature. No hint is given that these nobles commanded
vast personal clientelae that might determine the outcome of an election;
the value of nobility is measured in prestige not in clients. Discussing
the electoral prospects of Scaurus in 54, Cicero noted: `his father's
49
Comm. pet. 201; 38. In 65 Cicero had considered defending Catiline in court with
that advantage in mind: `Spero, si absolutus erit, coniunctiorem illum nobis fore in
ratione petitionis', Att. 1.2.
50
Comm. pet. 18: `. . . etiam si suffragandi studia non navant, tamen adferunt petitori
aliquid dignitatis', cf. 4.
Elections 109
memory counts with the country voters'.
51
In this case the advantage he
could expect from his ancestral background was directly bound up with
the personal standing of his father. Nobles, it seems, could not simply
deliver the vote by calling up their dependants. Votes had to be won, and
that was primarily done by courting the tribus. We nd no trace of large
standing armies of loyal clients under the command of individual noble
families. In fact lower-class clients are conspicuous by their absence
from our sources on late republican politics. The salutatio was probably
not a gathering of the poor and destitute. Neither were the assectatores
necessarily from humble backgrounds.
That in turn takes us to another aspect of campaigning; for how were
members of the lower classes activated in elections? The Commentar-
iolum pays very little attention to this issue, suggesting few direct
contacts between nobles and the lower strata of the plebs.
52
Indirect
connections may have existed, although that will have to remain hypo-
thetical. One way of targeting these groups was to inuence the leaders
of plebeian organisations; as we saw, Cicero is given this advice in his
brother's essay: `Then, reckon up the whole city all the collegia,
montes and pagi, neighbourhoods; if you strike up a friendship with the
leading men from among their number, you will easily, through them,
secure the masses that remain.'
53
We may doubt, however, whether their
recommendations would really have been sufcient to make the
members turn up for long exhausting days of often futile voting
sessions. In general it seems that the plebs was canvassed along exactly
the same lines as the elite, that is by granting favours which put the
beneciaries in one's personal debt. In the case of the plebs the favours
were primarily of a material nature: donations of sportulae, dinners,
games.
54
The practice of making donations to one's tribules was elevated as a
time-honoured tradition: as Cicero claimed, `These are the rewards and
bounties that poorer men receive from their fellow-tribesmen by ancient
custom', also noting, `There have always been ``good men'' who wanted
51
Att. 4.16.6: èst pondus apud rusticos in patris memoria'.
52
The plebs is explicitly dealt with only in paragraphs 30 and 32, while in 51 and 54 the
signicance of taking a popular stance is noted as a means of ingratiating oneself with
the plebs.
53
Comm. pet. 30: `Deinde habeto rationem urbis totius, collegiorum omnium, montium,
pagorum, vicinitatum: ex his principes ad amicitiam tuam si adiunxeris, per eos
reliquam multitudinem facile tenebis.' These organisations were used for political
purposes on a few other occasions, e.g. the recall of Cicero from exile, Cic. Dom. 74,
where the effect is likely to be grossly exaggerated.
54
On occasion these favours might be more `political' in nature. Thus Cicero in his
consular campaign may have tried to cash in on his previous support for Pompey who
was widely popular with the plebs, Comm. pet. 30; 51.
110 Elections
to be popular with their fellow-tribesmen.'
55
Special divisores were used
by the elite to distribute the handouts.
56
They are rst attested in the
early rst century, and soon became powerful players in Roman politics;
in 67 they successfully opposed the Lex Calpurnia which tried to curb
their activities. The inuence which the donor obtained among his
tribules by these means was an important trading object among the
senators. Thus, Cicero argued against imposing strictures which `will
forbid [our children] to court the respect and affection of their fellow-
tribesmen, or tell them that it is wrong for them to secure for their
friends the votes of their tribe, or to look for a like service from their
friends in their own elections'.
57
The value of nobles' support was to a
large extent based on the position of inuence they had established
within their tribus and centuria.
58
Thus, in Comm. pet. 56 bribery is
seen not as a threat to the bonds between patron and client but as a
danger to the loyalty of a tribus/centuria to its benefactors, its necessarii.
59
More widely targeted distributions which reached outside one's own
tribus also occurred but were regarded as bribery and banned. The
offence, it seems, was not caused by the involvement of money in the
electoral process. Thus, it was perfectly acceptable to cultivate small
sections of the electorate, that is one's own tribus, and establish a
powerful position here which could be used in support of others. The
point was that despite the de facto purchase of votes, this practice
remained within a framework of elite co-operation. By contrast, indis-
criminate and immoderate municence subverted the position of other
nobles in their respective tribus, and thus ultimately the traditional
system of power bargaining which ensured that no one rose to promi-
nence without being morally indebted to large sections of the senatorial
elite.
55
Mur. 72: `Haec homines tenuiores praemia commodaque a suis tribulibus vetere
instituto adsequebantur'; Planc. 44: `Semper fuerunt viri boni, qui apud tribulis suos
gratiosi esse vellent.'
56
Liebenam (1903) 1237.
57
Planc. 45: `. . . ne observent tribules suos, ne diligant, ne concere necessariis suis
suam tribum possint, ne par ab iis munus in sua petitione respectent'. A glimpse of
such tribal favours is found in Cic. Att. 2.1.9, from 60: `Favonius meam tribum tulit
honestius quam suam, Luccei perdidit.' Cicero had promised Favonius his tribus, the
Cornelia, and takes pride in the fact that he carried it with an even greater majority
than Favonius did his own. Lucceius, on the other hand, was apparently less inuential
in his tribus, which had turned Favonius down despite his endorsement.
58
Prospective candidates would cultivate their tribus with particular zeal. Comm. pet. 32
mentions supporters `propter suam ambitionem qui apud tribulis suos plurimum gratia
possunt', and special sodalitates of hopefuls were formed in the 60s to develop and share
tribal support, cf. Comm. pet. 18.
59
`Video nulla esse comitia tam inquinata largitione quibus non gratis aliquae centuriae
renuntient suos magno opere necessarios.'
Elections 111
The scope of donations must in general have been quite limited,
although the electoral impact might be considerable.
60
Banquets could
be quite large; still they accommodated only a very small part of the
plebs.
61
Deaths in the family were exploited to give lavish games, and the
aedileship also became a golden opportunity to show one's generosity.
These donations are often seen as a gesture towards the entire urban
plebs, but the capacity was quite inadequate. Only a relatively small
number could attend the games, and access was controlled through the
issue of tickets.
62
In general we are not dealing with unfocused or
indiscriminate municence. All donations, dinners, games and congiaria
could be carefully targeted at particular groups and tribes. Presumably
the prime beneciaries were the inuential rural tribes;
63
few favours
probably ever reached the bulk of the urban plebs, which had little say in
the elections.
By these gifts the donor established a short-lived relationship with the
recipients, focused on a specic issue: the immediate furthering of his
career and bargaining power. It relied on the expectation of favours
being repaid at the ballot box. However, after the introduction of the
secret ballot there was no way of checking how individual votes were
cast. Only the collective moral pressure on a tribus to endorse their
benefactor was there to ensure their support. This uncertainty may have
been instrumental in the emergence of another, more direct type of
bribery which rewarded voters only after the poll had been taken. That
may in fact have been one of the most important consequences of the
written ballot. This type of bribery became a signicant aspect of
elections in the rst century, when enormous sums were diverted into
the purchase of votes, despite repeated attempts to curb the practice.
64
Very little is known about the practicalities of bribery, but two types of
60
Municence, and games in particular, were often used to explain electoral success, cf.
Cic. Mur. 378; 40; 53; Fam. 2.6.3; Att. 4.16.6. Some of the most famous were
Caesar's in 65, Suet. Jul. 1011; Plut. Caes. 5.5; Dio 37.8; and Scaurus' in 58, Pliny
NH 36.11315. On the importance of munera in general Cic. Off. 2.579. Cf. Veyne
(1990) 21233.
61
In 70 Crassus gave an exceptionally large public dinner, served at 10,000 tables, Plut.
Crass. 12.2. On public banquets see Deniaux (1987) 299302.
62
Cic. Mur. 723, cf. Lintott (1990) 89.
63
Signicantly the audiences at games are often described as boni, cf. Cic. Pis. 65; Sest.
115.
64
See e.g. Linderski (1985); Deniaux (1987); Lintott (1990) 810; Wallinga (1994); on
bribery in general see most recently Jehne (1995), who regards the practice simply as an
aspect of traditional euergetism: the distribution of small, almost symbolic, tokens of
the candidate's goodwill. That interpretation, while avoiding the moralisation of many
modern scholars, fails to explain how bribery could have had any signicant impact on
Roman elections, and thus why candidates were willing to spend huge sums on these
`symbolic' gifts. Nadig (1997) adds little new.
112 Elections
donations can be distinguished. The tribal congiaria were made before
the elections in the expectation that people would feel obliged to return
the favour on polling day. This type was a natural extension of the other
gifts made to one's tribules. The alternative form of bribery was more
blatant: the money promised was to be paid only in the event of a
successful outcome. It thus established a much more direct link between
support and remuneration than did the other system, which still worked
along the principles of patronage and the exchange of benecia and
ofcia.
None of these types of bribery could have included the entire urban
electorate, let alone the whole citizen-population. For sheer economic
and practical reasons that would have been unfeasible, and politically
such indiscriminate generosity was pointless. It is inconceivable that
more than 200,000 voters the urban citizen-body plus those rustici and
municipales who might have come in for the elections could have been
individually bribed. Moreover, considering the level of participation
possible in the comitia, it would make no sense to bribe the entire
electorate, the overwhelming majority of whom would not turn up
anyway. And if they did, most of them would have little impact on the
outcome. Thus, the idea of general bribery is effectively undermined by
the lack of inuence of the urban lower classes, the proletarii in
particular.
Our sources often refer to bribery of the `people' or the `vulgus', but
that is too unspecic to prove the plebs was bribed in its entirety.
65
In
some cases this possibility can be positively ruled out. Asconius says that
Milo, running for the consulship in 52, òpenly presented voters in the
tribes with 1,000 asses each'.
66
Despite its seemingly general character,
a quick calculation is enough to rule out this possibility. The expenditure
involved in bribing the entire urban population would have amounted to
more than HS 50 million. Not only are such outlays unrealistic, the
political returns of the investment would have borne no reasonable
relation to the expense involved. Yakobson has suggested that Milo's
bribe was unusually high, comparing it with Caesar's posthumous gift of
HS 300 to each member of the urban plebs, which may have accumu-
lated a total cost of perhaps HS 60 million.
67
However, apart from
leaving the political rationale unexplained, Yakobson's argument ignores
the possibility that Caesar's donation may have been exceptionally
generous precisely because it in contrast with previous sportulae
65
E.g. Cic. Att. 4.17.4: `populo tributim domi suae satis fecerat.'
66
33C: àperte quoque tributim in singulos milia assium dederat', cf. 35C: `. . . popu-
loque tributim singula milia aeris ad defendendos de se rumores dedisse'.
67
Yakobson (1992) 42; Millar (1998) 203.
Elections 113
included the entire urban population.
68
Milo's bribe must have targeted
particular sections within the plebs in the same way as games tickets and
presumably congiaria did, that is, with a strong focus on the rural tribes.
The most common practice of bribery appears to have taken the form
of promises, whose fullment was made dependent on electoral success.
The predominance of this type is indicated, for example, by the law
proposed by the tribune Lurco in 61. According to Cicero, he suggested
that àny person promising money in a tribe shall not be punishable
provided he does not pay it; but if he does, he shall be liable for HS
3,000 to every tribe for life'.
69
Several other passages also refer to the
promise of bribes.
70
The use of sequestri, middlemen who kept the
promised sums until the election was over, also appears to have been
common.
71
It seems that bribes were usually promised at a specied rate
in advance of the election; the money was deposited with sequestri, and
in the case of a successful outcome it was then distributed by the
divisores, who were attached to the individual tribes.
The promise cannot have covered voters in general. If it had, tens of
thousands of citizens who had not taken part in the vote would have
been able to turn up afterwards and claim a share of the reward. On the
other hand, it is also difcult to see how it would have been possible to
pay only those who had actually voted. There was no formal identity
check of the voters and no register of their attendance. The divisores, if
present at the assembly, might have kept an informal register. But that
could hardly have been kept a secret, and would, if noticed, provide
incriminating evidence in later ambitus trials, where we never hear of
divisores checking the attendance of voters. Alternatively the divisores
might simply have remembered who had turned up. Although that
might have been possible in the rst class or in very small tribus, that
solution must be ruled out for the populous urban tribus. The sheer size
of these tribes represents a fundamental obstacle to general bribery; for
68
Caesar's distribution was organised by the curatores tribuum, not the divisores, App.
B. Civ. 3.88. The curatores probably kept records of all the tribules, cf. Dion. Hal. 4.14.
Possibly the networks maintained by the divisores were much less comprehensive. It
seems unlikely, for example, that they should have been in direct contact with members
of the urban proletariat. In 60 Balbus left HS 100 to each citizen in Rome, perhaps
around HS 25 million, Dio 48.32.
69
Att. 1.16.13: `. . . ut qui nummos in tribu pronuntiarit, si non dederit, impune sit, sin
dederit ut quoad vivat singulis tribubus HS 3,000 debeat'.
70
Suet. Div. Jul. 19.1; Cic. Planc. 45. Cic. Cluent. 75 does not, as claimed by Yakobson
(1995) 439, show that bribes were generally paid in advance. The `nummos suppressos'
are more likely to be money promised but not paid, cf. Cluent. 71: `. . . pecuniam . . .
polliceatur, deinde eam postea supprimat'.
71
Cic. Verr. 1.36; 2.2.108; 2.44; Cluent. 25; 72; 87; Planc. 38; 478; Cael. 16; 30; Comm.
pet. 57; Cic. ap. Asc. 83C.
114 Elections
how did one avoid the bulk of urban tribesmen turning up and comple-
tely draining the coffers of the campaign? The comitia centuriata pre-
sented particular problems. Here the lower classes were not organised in
separate tribus, but voted mixed. Even if the tribal organisation also
applied here, as seems likely, several tribes would combine in one
centuria, making it difcult for the divisores to check who had turned up
from their respective tribus.
72
However, we know that promises could be made to all the tribes. One
example comes from 69, when Verres tried to prevent Cicero from
reaching the aedileship, Cic. Verr. 1.223. He invited divisores from all
the tribus to his house, offering them the commission to distribute
HS 500,000 to obstruct Cicero's election.
73
The involvement of all
tribus, however, does not imply that all tribules were promised money.
The rural tribus must have been favoured to the near exclusion of the
mass of urban tribesmen. The question is how this was done in practice,
bearing in mind that there was no way of keeping records of voters, and
payments to the mass of urban tribules had to be avoided. The evidence
from the case of Plancius may provide a clue.
He was charged with bribery after the aedilician election of 54. More
specically the trial revolved around the accusation that Plancius had
been a member of a sodalitas and conducted `decuriatio' and `con/
inscriptio tribulium', been a sequester, promised bribes and distributed
them, Cic. Planc. 45; 47.
74
The `conscriptio' and `decuriatio' of tribules
were explicitly linked to the bribery, and unrelated to the spurious
clause on vis attributed to the Lex Licinia de sodalitatibus, under which
Plancius stood accused.
75
`Decuriare' simply means to divide into
smaller groups, while `conscribere' refers to the act of taking down
names on a written list. This information thus offers a unique glimpse
into the organisation of bribery. In Plancius' case the bribe was alleged
to have been promised in advance, Planc. 45. The lists of tribal voters,
indicated by Cicero, would therefore have been drawn up before the
election by agents involved in the bribery scheme. By specically
targeting individual voters, who were personally enlisted, it was possible
to organise bribes which were made dependent on the outcome of the
72
Cf. Yakobson (1995) 435. On the practicalities of bribery see also Jehne (1995) 723.
73
Cf. Cic. Har. resp. 42, also refers to `divisores omnium tribuum'.
74
The allegation against Plancius that he had been a sequester may seem surprising
considering the fact that he was prosecuted for bribing voters at his own aedilician
election. He was, however, charged with the crimen sodalitatis under the Lex Licinia, and
not ambitus in general. Cicero is therefore not describing the specic acts linked to
Plancius' own bribery of the tribes in 54, but the wider range of operations involved in
the membership of a sodalitas.
75
Elections 115
election. The voters in question were promised a reward for their
support, and if the candidate succeeded, their remuneration would be
paid presumably under the proviso that their particular electoral unit
had come out in favour. This system did not rely on actually checking
their vote or attendance, but on the voters' personal interest in the
candidate winning in their centuria or tribus. The semi-contractual
relationship between candidate and voter may explain the dissatisfaction
which erupted in cases where the candidates did not full their prom-
ises, or there were disputes over the exact terms, cf. Cluent. 75.
76
By this practice the huge gap between the size of the citizenry and that
of the active electorate could be overcome. By limiting the offer to
particular voters whose unit could be checked, it was possible to
promise bribes to an electorate whose vote or attendance could not be
individually checked. It was a tight and rational way of organising tribal
bribes, which also allowed the outlays to be effectively controlled; a
candidate simply enlisted as many voters as he could afford at the
chosen rate.
The importance and electoral impact of bribery in the late republic are
indisputable. It was most widespread in the consular elections, which
raises the classic problem of explaining how bribery could be of any
consequence in the timocratic centuriate assembly where the poorer,
supposedly more `corruptible' voters had little inuence and the proper-
tied classes were able to carry the vote on their own. The lower classes
may often have had a say in the elections, and they could obviously be
bribed by relatively small outlays. But without strong support from the
rich, the lower classes, no matter how effectively bribed, could not
deliver the goods. Bribing the rich was obviously expensive, and the few
reported sums may not indicate a `standard level' of bribes. Presumably
the 250 sesterces distributed by Milo would have had little impact on
voters of the rst class, where the rates paid must have been much
higher. In 54, 10 million sesterces were promised to the praerogativa
centuria. Assuming an attendance of, for example, 150 voters in a rst-
class centuria, that would make a donation of no less than HS 16,000 for
each voter. The scale of expenditure was clearly enormous that year it
even led to a rise in interest rates and by clever targeting the effect
could be optimised.
77
The secret was to aim at rst-class centuriae of
very small or poor tribus which had only a few members in the rst class.
76
Problems might for example arise if a candidate, despite overall success, had lost in
some units, where the bribed voters still insisted on receiving a share in the
remuneration.
77
Cic. Verr. 1.26, and Schol. Gronov. 350 (St), also suggest that promises to the
praerogativa centuria were not uncommon.
116 Elections
In those cases even smaller sums might give considerable electoral
returns. The advantage would be further enhanced if, as seems likely,
the bribed voters only voted for one candidate, bringing him even
further ahead than a vote for a full magisterial college would have done.
Parallel to these attempts to win the support of individual voters,
candidates also canvassed the public in general. In this part of the
campaign the objective was to create a favourable impression, raise one's
public prole, and cultivate an image of personal success. In itself this
would hardly have brought many voters to the ballot-boxes, but it might
prove important in securing `second' votes from people who were
otherwise committed. Although the people who turned up for an
election would have done so out of personal obligation to one of the
candidates, the electoral importance ascribed to public appearance also
suggests that voters were sufciently independent to dispose freely of
their remaining votes.
The public at large was courted by a variety of means. Carefully
orchestrated public appearances played an important part. When a
candidate went out, his personal popularity and social standing had to
be emphasised by a large entourage, which would never leave his side.
Extra sectatores might even be hired to enhance the effect, Cic. Mur.
701; 73. Candidates had to put on a show, as the Commentariolum
advises: `Lastly, see that your whole canvass is a ne show, brilliant,
resplendent, and popular, with the utmost display and prestige.'
78
In
presenting oneself to the public certain advantageous features, like
military prowess, forensic fame or illustrious ancestry, would be duly
stressed. Previous municence would also have been invoked at this
stage.
79
Personally the candidate had to guard his expressions very
carefully in order to present both a serious and a forthcoming gure to
the public, cf. Comm. pet. 44. A single ill-considered remark might travel
and damage his campaign.
80
That particular hazard brings us to another aspect of public cam-
paigning: the attempts at gossip and news management.
81
In order to
ensure that he was well spoken of, the candidate had to ingratiate
himself with his neighbours, whose personal experience of him might
78
Comm. pet. 52: `Postremo tota petitio cura ut pompae plena sit, ut inlustris, ut
splendida, ut popularis sit, ut habeat summam speciem ac dignitatem', cf. 41.
79
Cf. Cicero's reference to Scaurus and Domitius in Att. 4.16.6.
80
Thus the story in Val. Max. 7.5.2, about the failure of P. Scipio Nasica, cos. 111, to win
the aedileship. Noticing the rough hands of an agrestis, he asked whether he walked on
his hands. The joke was picked up by those standing around him and àd populum
manavit causamque repulsae Scipioni attulit'.
81
On the candidates' fear of rumours also Cic. Mil. 42: `. . . sed etiam quae obscure
cogitari possunt timemus, rumorem, fabulam falsam, ctam, levem perhorrescimus,
ora omnium atque oculos intuemur'.
Elections 117
inuence public perception. For the same reasons his own household,
including slaves, should be treated with great consideration in the run-
up to an election. Otherwise hostility from those closest to the candidate
might lead to details of his personal life being leaked and damaging his
reputation. As Quintus Cicero reminded his brother: `the talk which
makes one's public reputation generally emanates from sources in one's
own household'.
82
The way the general campaigns were conducted further underlines
the àpolitical' nature of the elections. Though potentially the most
`political' type of campaigning, they were focused entirely on personal
qualities or vices. Political issues, it seems, were avoided at all costs;
and not simply those that might cause controversy, but issues in general.
Certain inuential groups, like knights, publicani or Italian nobles, might
be singled out for particular praise, but not as part of a specic policy or
at the expense of other social categories. It follows that political speeches
had no place in campaigning and were largely irrelevant to the pursuit of
a public career. Cicero, for example, made his rst public address in 66
when he was already high up on the magisterial ladder, holding the
praetorship, Manil. 13. On this occasion he came out in support of
Manilius' proposal to grant the Mithridatic command to Pompey. But
however issue-related this speech might seem, it was above all an
attempt at gaining Pompey's support for his consular candidature,
incidentally also exploiting Pompey's popularity among the plebs, cf.
Comm. pet. 5; 51.
This chapter has made an attempt to dene the nature of popular
participation in elections. The independence of the voters has been
stressed as a crucial feature which must be taken into account in any
reconstruction of social and political life in the later republic. This
element is strongly suggested both by the extensive campaigning by the
elite and by the erratic and spontaneous voting patterns observed in the
assemblies. However, to construe this freedom of action as a `demo-
cratic' quality misses the peculiar character of the Roman elective
assembly. Firstly, the inuence of the sovereign people was heavily
circumscribed, and large-scale participation actively discouraged by a
number of factors. Secondly, the elections had little political substance
that might attract popular interest and persuade the plebs to overcome
the obstacles raised against its participation. Thirdly, it follows
from these points that plebeians had no natural presence in elections,
where their limited attendance must be explained as a result of
mobilisation by the elite. Finally, despite the importance of planning
82
Comm. pet. 17: `Nam fere omnis sermo ad forensem famam a domesticis emanat
auctoribus.'
118 Elections
and organisation by the elite, the outcome remained unpredictable due
to the size of the electorate, the political independence of the voters, and
not least the multiple vote, which added even greater uncertainty to the
process.
The conclusion, summarised here, applies almost exclusively to the
last century BC, from which most of our source material derives. This
period was in many respects an extreme age, characterised by social and
political upheaval, and some scholars have identied the extension of
the Roman citizenship to the Latins and Italian allies in the wake of the
Social War as a prime factor behind the decline in the political culture
during this time.
The enfranchisement of the Italians was the single most dramatic
change to the Roman electorate in the late republic. When nally
implemented by the census of 70, the result was a doubling of the
citizenry, which rose to more than 900,000. Historically this was a
change of tremendous importance, but its effect on the practice of
politics in Rome is not altogether clear. The supposed Italian demand
for Roman citizenship has been widely seen as an attempt to gain
political inuence in Rome, in which case the consequences might have
been substantial.
83
Thus, it has been suggested that the entry of
hundreds of thousands of uncommitted voters into the citizen-body
seriously upset the traditional patterns of politics, also leading to an
increase in bribery.
84
First we will have to ask to what extent the new citizens participated
in the assemblies. Since the communities closest to Rome with few
exceptions had already been enfranchised before the Social War, most of
the new citizens would have lived too far from the capital to be able to
participate on any regular basis. Not surprisingly the evidence for extra-
urban participation is very limited. Cicero mentions: `these crowds of
people from the whole of Italy . . . that have simultaneously assembled
because of the elections, the games, and the census'.
85
The census of
70, to which Cicero refers, was the rst one to be remotely effective
since the formal enfranchisement of the Italians in the 80s, and the
turnout that year may therefore have been exceptional.
86
The passage
83
For a different interpretation of the Social War and its political and cultural background
see Mouritsen (1998).
84
Wiseman (1969) 657, cf. Paterson (1985), esp. 27; Millar (1998) 211.
85
Verr. 1.54: `haec frequentia totius Italiae . . . quae convenit uno tempore undique
comitiorum, ludorum censendique causa'.
86
It has been argued that for practical reasons the census in 70 must have been conducted
locally. Cicero's formulation in Verr. 1.54 may also call into question the idea that all
citizens had to turn up in Rome for the census. Thus, if all the 900,000 citizens
registered in 70 had appeared personally before the censors in Rome (in itself involving
quite unrealistic logistics of travel, accommodation and the actual registration), it
Elections 119
does not therefore in itself prove that large numbers of Italians regularly
attended the assemblies. Murena received support from Umbria, Cic.
Mur. 42, and so did Cicero from Volaterrae and Atella, Fam. 13.4.1;
13.7.4. Another case comes from 54 when Cn. Plancius' aedilician
candidature apparently enjoyed strong support from his native Atina
and the neighbouring towns of Arpinum, Sora, Casinum, Aquinum,
Venafrum and Allifae, Cic. Planc. 22. In his speech for Plancius Cicero
speaks of large numbers turning up, but we have no idea what consti-
tuted a large crowd in this context. Moreover, it remains a big question
whether Plancius' local friends would otherwise have turned up for the
vote.
Plancius' prosecutors seem to have used his background as an eques of
Italian stock to substantiate the allegation of bribery, claiming that only
by fraud and bribery could a man of his humble origins have beaten a
Roman nobleman, Planc. 1718. Cicero turns this argument around,
arguing that it was precisely Plancius' Italian roots which had helped
him secure the aedileship, 1923. Thus, the victory is explained by the
support he received from his local region in central Italy. However, if we
look at the tribal allocation of the towns listed by Cicero, we nd that
apart from Sora's tribus, the Romilia, they were all quite large and often
split into several geographical sections.
87
Cicero nevertheless argues that
the support raised in Plancius' neighbouring towns was able to carry the
tribes in his favour. The implication would be that the turnout from the
other towns inscribed in these tribes had been negligible. One tribus, the
Volturia, Plancius apparently controlled so effectively that at some point
before the election he could promise it to Laterensis; eventually,
however, he aligned himself with Plotius, with the result that Laterensis
only received a handful of votes there. But while the Volturia may have
included the neighbouring town of Audena, it also covered more
remote parts of Samnium, Lucus Feroniae and perhaps Castrum
makes no sense to list the census simply as one among several other reasons for the
inux mentioned after the comitia and the games; the order of magnitude would have
been entirely different. On the other hand, Cicero's reference to people arriving from
all Italy `censendi causa' is unequivocal. A possible explanation might be that the
Italians in question had not come to register, but were local census ofcials presenting
their gures to the censors in Rome. These magistrates would themselves have
numbered almost a thousand, and assuming they were accompanied on this important
mission by a substantial entourage of friends, relatives and servants, they would have
made a strong contribution to the mixed `frequentia', mentioned by Cicero, especially
if their arrival had been timed to coincide with games and comitia. Thus also Millar
(1998) 28.
87
The tribus of Arpinum, the Cornelia, was split into numerous geographical parts, as was
the Oufentina, the tribus of Aquinum. Casinum, Venafrum and Allifae all belonged to
Plancius' own tribus, the Teretina.
120 Elections
Novum, in which Plancius would probably have had few or no direct
contacts. The implication is that these towns had been very poorly
represented in the aedilician election of 54.
It seems that the tribules from rural districts mentioned in the Pro
Plancio had turned up for the election only because a local noble ran for
ofce and had called on their support. Therefore, instead of showing the
political importance of the new citizens, Plancius' case would tend to
underline their general absence from at least minor Roman elections.
88
The suggestion that Italians did not take part on any regular basis
does not necessarily imply that they were politically unimportant. The
overall scale of participation meant that even a limited turnout from the
towns of Italy might have a real impact on the elections. Considering
the number of towns, estimated at around 400, it would take little more
than a few handfuls of decurions from each to carry the vote in the
comitia centuriata.
89
It could therefore be argued that the disadvantage of
geographical distance to some extent was compensated for by the scale
and structure of the assemblies, which favoured rural and propertied
voters. For those reasons campaigning in Italy might still have been a
worthwhile effort. The question is whether Roman candidates exploited
this opportunity to any signicant extent.
The evidence for extra-urban canvassing is dominated by the example
of Cicero, whose bid for the consulship in 63 remains the best-
documented campaign from the late republic. As early as 65 he had
planned a trip to Cisalpine Gaul; the purpose was to promote his
candidature in a region about which he noted that it `seems to be able to
carry much weight in the polls'.
90
A later reference in Phil. 2.76 might
suggest that Cicero actually went there. In the Commentariolum Cicero
88
Cicero, Vat. 36, claimed that Vatinius had lost his own tribus, the Sergia, in the aedilician
election of 58 because of the rejection by his rural tribesmen. His description of them,
however, as `severissimorum hominum Sabinorum, fortissimorum virorum Marsorum
et Paelignorum, tribulium tuorum' suggests that Cicero's emphasis on the rural vote was
an example of the common topos of the morally astute Italian peasant rather than a
factual reference to their dominance in the comitia, cf. in general Dench (1995). Cicero
simply seized the opportunity, offered by Vatinius' defeat, rhetorically to contrast the
virtues of these idealised `montani atque agrestes' with the depravity of Vatinius. It does
not therefore prove that these voters maintained a regular presence in the elections.
Taylor (1960) 263, suggested that Vatinius' family hailed from the Marsic region, which
would of course have added extra poignancy to his failure to mobilise their support.
89
Based on the gures of Beloch, Lo Cascio (1994) 37, puts the number of towns in
Augustan Italy at 434, 380 of which were located in Cisalpine Gaul and the peninsula.
90
Att. 1.1.2: `. . . videtur in suffragiis multum posse Gallia'. Shackleton Bailey
(196570) I, 127 translates the passage: `Gaul looks like counting heavily in the
voting', but that does not express the potential element in Cicero's formulation,
presenting as it does the inuence of Cisalpine Gaul as a well-established fact in
contemporary politics.
Elections 121
was famously advised to learn the tribal map of Italy by heart,
91
and the
treatise also describes how the support of the domi nobiles could be won.
As in Rome, it was a question of using favours and obligations, support
being promised to ambitious municipales and demanded from those who
had already received favours. Later Cicero triumphantly proclaimed
that `cuncta Italia' had turned out for his election, e.g. Pis. 3.
Cicero's case might suggest that the Italians had become important
players in Roman elections and were courted with considerable energy
by the candidates. The question is, however, whether Cicero was really a
typical candidate from whose example general practice can be described.
His municipal background obviously made him an exception among
consular candidates; most likely it also forced him to campaign with
much greater determination. There are signs that Cicero put unusual
effort into courting the Italian constituency. It appears from the Com-
mentariolum that he had come to know the domi nobiles much better than
his competitors, and a special relationship with this group is suggested
by his claim of general support from `tota Italia' in the election,
92
and
not least by the strategy used to bring him back from exile. Exception-
ally, the comitia centuriata was called and Italian voters were mobilised to
pass the bill. This Italian emphasis is apparent also in Cicero's frequent
elevation of the virtues of the Italian peoples and their elites, which were
presented as the moral bedrock on which the republic could be rebuilt.
In these efforts we may distinguish the outline of a particular Italian
agenda, by which Cicero tried to turn his inferior personal background
to his advantage, using it as a platform on which an independent power-
base outside Rome could be founded.
93
In that case it follows that his campaign may not have been at all typical,
devoting an unusual level of attention to extra-urban voters. Generally
we have conspicuously few traces of electoral canvassing outside the
capital. The oft-mentioned instances of campaigning in Cisalpine Gaul
are highly suspect as proof of regular canvassing in this region.
94
And
91
Comm. pet. 30: `Postea totam Italiam fac ut in animo ac memoria tributim discriptam
comprehensamque habeas . . .'
92
Although Sallust claimed that Cicero owed his consulship to the support of the nobility,
Cat. 23.6, the Italian backing may well have been an important factor. Thus, in 59
Cicero anticipated strong Italian support in case Clodius took action against him, Q.
Fr. 1.2.16: `tota Italia concurret'.
93
Thus, at the end of the Commentariolum, 58, Q. Cicero stressed that the advice given in
the essay was not of general relevance but applied specically to his brother's situation.
94
Cicero probably went on a campaigning trip to Cisalpine Gaul but, as argued above, he
may have been exceptional in his assiduous canvassing of the Italians. Moreover, his
comment on the electoral inuence of the northernmost Italian region, Att. 1.1.2, was
probably linked to the favourable tribal allocation of the region rather than a reference
to crowds of Cisalpine Gauls regularly lling Roman assemblies.
According to Hirtius, BG 8.50, Caesar travelled to Cisalpine Gaul in 50 in order to
122 Elections
while Plancius simply mobilised support in his home region, Murena
cashed in on past favours offered to the Umbrians presumably he
simply wrote to them and requested their attendance. The only known
examples of candidates canvassing outside their home region come from
the immediate surroundings of Rome. In 52 Clodius, then a praetorian
candidate, was killed on his way back from an electoral meeting with the
councillors of Aricia, Asc. 31C. We also know from Cicero that Plancius
had given games in Praeneste, Planc. 63. This pattern of campaigning,
concentrated in and around Rome, makes good sense considering the
scale of the assemblies and the structure of the tribal map. Only a small
number of voters was required to carry a tribus or centuria; if they could
be found in the vicinity of Rome there was no point in extending the
campaign much further, to areas where the electoral returns presumably
would have been relatively much lower because of the geographical
distance. Thus, for a candidate seeking support from, for example, the
tribus Horatia, it would have been both easier and more protable to
campaign in Aricia than going all the way to Venusia in Apulia.
Presumably these voters could be safely ignored since so few of them
would turn up anyway. Their general absence could explain why
Cicero's competitors for the consulship apparently had neglected the
Italian constituency during the campaign of 64; Quintus notes of the
Italians that `To the rest, especially to your competitors, they are total
strangers.'
95
The Italians, it seems, could be ignored for the simple
reason that they, like most other Roman voters, had no regular presence
in the assemblies. Later Cicero complains about the apathy of the Italian
elites, who were more interested in their `villulas' than in the fate of the
republic.
96
Apparently the Italians did not become a political force
raise support for Antony's bid for the augurate. When he realised that this election had
already been successfully fought, he decided to stay and canvass for the forthcoming
consular election. The campaigning area was obviously determined by the limits of
Caesar's province and the fact that he could not have canvassed anywhere else. The
story is generally suspect. It seems unlikely that Cisalpine Gauls, or anyone else, would
at their own initiative have gone to Rome simply to vote on the augurate, which was
decided by only seventeen tribes selected by the drawing of lots. As suggested by
Bleicken (1975) 256, Caesar may in fact have been mobilising veterans whom he
provided with money for the journey to Rome, a method already used at the consular
election in 55, Dio 39.31.2; Plut. Caes. 14.6; Pomp. 51.4.
In Phil. 2.76 Cicero implies that Antony canvassed in Cisalpine Gaul for the consular
election in 45. That is quite implausible, given the fact that candidates were then
brought forward by Caesar in numbers matching the posts to be lled. Therefore,
whatever Antony was up to in Gaul, it cannot have been electoral campaigning in any
traditional sense.
95
Comm. pet. 31: `Hos ceteri et maxime tui competitores ne norunt quidem . . .'
96
Att. 8.13.2; 8.16.1. In Att. 2.6.2 Cicero also suggests that the burghers of Antium had
little interest in urban politics.
Elections 123
which every Roman noble seeking ofce had to reckon with and pay due
attention to.
We have therefore no compelling reason to believe that the Italian
enfranchisement changed the pattern of elections or the nature of
popular participation. The candidates probably tried to mobilise ex-
isting contacts outside the capital, urging friends and acquaintances to
turn up for their election. But there are no traces of Roman candidates
systematically canvassing this constituency. Italian enfranchisement
meant an enlargement of the electoral pool which again opened up new
opportunities for some candidates to explore, most notably Cicero who
seems to have pioneered this alternative strategy. The sources do not
allow us to say to what extent his example was followed by other
candidates.
97
It seems clear, however, that Rome did not experience any
substantial or instantaneous inux of new, volatile voters into the
assemblies. The numbers were small and mobilisation from Rome
remained essential.
The methods used to raise Italian support were no different from
those employed in the capital. Again `spe utilitatis' was a central feature,
Comm. pet. 312. The ambitions of Italian nobles were exploited, and
local backing traded against future support. Mobilisation probably
worked selectively, targeting individual towns where a candidate had
personal contacts, whom he urged to turn up with a loyal following. As
noted, Cicero explicitly mentions support from Volaterrae and Atella,
towns with which he had already established special links.
98
Therefore,
although the numbers involved may have remained limited, the en-
franchisement of the Italians is likely to have raised the general level of
attendance in the elective assemblies.
In this respect the new Italian citizens may have contributed to a
general trend in late republican politics, when popular participation was
stimulated by a number of new factors. There was an obvious increase
in bribery, and the methods used were constantly rened. The scale of
municence rose dramatically, and a new feature was introduced with
the mobilisation of veterans as electoral backers.
99
They probably all
had the effect of drawing more people into the electoral process. Viewed
97
Remarkably few Italians seem to have entered the senate before the end of the republic,
which might suggest a rather limited integration into the urban networks of patronage
and electoral campaigning. Contra Wiseman (1971) 36, who appears to overestimate
the representation and inuence of the Italians.
98
Cicero had defended Caecina, a member of the leading family in Volaterrae, and was
the patron of Atella, Q. Fr. 2.13.3. Likewise, Murena in his consular election exploited
the gratitude he had earned in Umbria by conducting a favourable levy there, Cic. Mur.
42.
99
E.g. Cic. Att. 4.16.6; Mur. 378; Dio 39.31.2; Plut. Crass. 14.6; Pomp. 51.4.
124 Elections
in this perspective the most salient feature of rst-century political
participation may not be the supposed breakdown of traditional stan-
dards and loyalties but rather an overall increase in the scale of the
process.
This development had already begun in the previous century, where
we nd the rst signs of a new, more competitive form of electioneering.
Livy, 37.57.11, refers to generous congiaria distributed by new men in
189, and the rst half of the second century saw a spate of legislation,
regulating electoral competition and ofce-holding. The Lex Baebia de
ambitu was passed in 181 and the rules tightened again in 159, Livy
40.19.11; per. 47, before the Lex Cornelia Fulvia de ambitu was intro-
duced in 149. The games organised by the aediles were also regulated in
179, Livy 40.44.1012. Whether bribery and electoral malpractice were
quite as rife as these initiatives might suggest is an open question,
however; as Develin noted, they may have been attempts to nip an
incipient problem in the bud.
100
Still, there are distinct signs of a change in the political climate in this
period. In 166 the elections were conducted àmbitiosissime', Obs. 12,
and Polybius described what was apparently a new form of canvassing.
In a discussion of Scipio Aemilianus' early career he noted, 31.29.89,
that: àll the time that other young men gave up to law affairs and
greetings, spending the whole day in the Forum and thus trying to court
the favour of the populace', Scipio devoted to hunting, which was an
occupation Polybius regarded as more tting for a young noble than the
demeaning pursuit of popular favour. The impression is one of a fairly
recent development, which was looked upon with considerable unease
among the ruling circles. After the Hannibalic War it seems that a more
competitive political climate developed. The electoral campaigning
described in book 31 may have been identied as the rst signs of the
inevitable political decline, which Polybius predicted would follow the
expansion of any state. Thus, in 6.57.59 he envisaged that as life
becomes more extravagant: `citizens will become more erce in their
rivalry regarding ofce'. The struggle for high ofce naturally intensi-
ed, as these themselves grew in attraction and protability. Ultimately,
therefore, the changes to the elections reected much broader historical
processes, that is the growth of the Roman empire.
The Roman elite had probably always competed for public ofce.
101
100
Develin (1985) 31415.
101
Contra Develin (1985) 30915, and Wiseman (1994) 329, who speaks of `newly
competitive, newly unpredictable elections' in the late republic. See however,
Rosenstein (1993) and Pani (1997) 1967, who demonstrates convincingly that
already in the early second century no pattern or predictability can be traced.
Elections 125
But traditionally the collective ethos of the ruling class had been strong
enough to ensure that ofce-holding and thus recruitment to the
senatorial order was managed with a minimum of popular involve-
ment. As we have seen, the Roman assemblies were never promoted as
vehicles of `democratic' representation. Viewed in that perspective the
stable rule of the aristocracy and its monopoly on high ofce no
longer presupposes a comprehensive system of personal control, which
tied each voter to a member of the elite. The `popular' element could be
effectively sidelined by a consensually based code of conduct which
restrained elite campaigning and limited popular mobilisation in the
run-up to the elections.
It follows that the system did not break down due to popular demands
for greater inuence and choice. It did so when the elite could no longer
contain the growing dynamics within its own ranks. Increased competi-
tion undermined the senate's ability to manage the fundamental conict
between, on the one hand, its collective interest in controlling an
unpredictable and potentially disruptive popular institution and, on the
other hand, the pressure on its individual members to raise greater
personal support in order to reach ever more covetable and alluring
public ofces.
Vain attempts to reinforce a collective discipline were made through
the introduction of stricter rules regulating campaigning practices and
ofce-holding. Clearly, the aim was to contain the electoral competition
and regulate access to public honours rather than to ensure a
democratically sound procedure. The senate obviously had no interest
in allowing the will of the people to be expressed as freely and directly as
possible, and ambitus laws were therefore not concerned with electoral
malpractice in the modern sense. Roman legislation only targeted
certain types of payment to voters; on a limited tribal level the distribu-
tion of favours and money remained fully acceptable. Likewise, the ban
on other, seemingly innocent means of canvassing such as the use of
nomenclatores and sectatores, suggests that the aim was to curb candi-
dates' general ability to mobilise voters.
102
Evidently, bribery laws also
sought to reduce the spiralling costs involved in electioneering, creating
a more level playing-eld for the nobles involved. Thus, the ultimate
objective was to maintain a pluralistic system of elite co-operation,
based on the exchange of favours and the sharing of inuence under the
collective authority of the senate.
102
The lex Tullia de ambitu (63) also instituted a biennium, a two-year period in which
candidates were barred from giving gladiatorial shows èxcept in execution of a will or
on a date xed beforehand', Crawford (1996) 7612.
126 Elections
Any attempt to halt the process was doomed, however. The senate
was up against historical forces far beyond its control. The conquest of
the Mediterranean world had raised the stakes of magisterial elections so
high that the potential benets from ofce-holding now outweighed any
risk of sanctions. Sulla's expansion of the senate may for a period have
eased the pressure to enter the highest order. But the increase in the
number of lower ofces merely narrowed the bottleneck to reach the
higher, more lucrative and prestigious posts. Moreover, the censors'
expulsion of sixty-four senators in 70 may have further increased
competition at the lower levels, as those expelled tried to regain their
status.
103
At this time the senate's authority had already weakened, and
the collective interests of the elite could no longer be effectively safe-
guarded. The traditional strategies to limit popular participation were
undermined by the internal dynamics within the elite, which forced its
members to campaign harder and more widely than before. The in-
creased efforts to mobilise voters broke the barrier which had tradition-
ally existed between the electoral process and the daily lives of most
citizens.
Republican Rome was in the grip of constant electioneering; every
year was an election year with no fewer than forty-four `political' posts
up for reappointment, in addition to fty lower ofcials.
104
The number
of candidates for each of these is unknown, but already in the second
century there were ve to seven contestants for the consulship, and in
64 no fewer than ten.
105
Presumably every post was contested in this
period, and for each candidate electoral success was crucial to his future
career and prospects. As competition intensied, the rules imposed to
regulate the process were increasingly outed. The result was not a
`politicisation' of the elections, which remained deeply personalised
contests rarely involving political issues of general interest. Still, the
extensively conducted campaigns of the late republic, often planned
years in advance, did have a profound effect on life in the capital; games
were given more frequently and on an ever grander scale, tribal dona-
tions were made, bribes distributed, and individual voters approached
and canvassed. The social stratum most affected was not, however, the
broad population, which probably stayed on the margins of the electoral
process. Those who felt the changes to the political climate were
103
On expulsions from the senate in general see Evans (1997).
104
`Political' posts: 2 consuls, 8 praetors, 4 aediles, 10 tribuni plebis, 20 quaestors; lower
ofcials: 24 tribuni militum, 26 vigintisexviri, including tresviri capitales, tresviri
monetales, decemviri stlitibus iudicandis, praefecti Capuam Cumas, cf. Kunkel (1995)
53251.
105
Jehne (1995) 54.
Elections 127
primarily the propertied classes, the boni, who represented public
opinion and had a real say in the assemblies. Occasionally wider circles
of the urban population may have beneted, but those cases probably
remained exceptional.
6 Plebs and politics
Having dealt with the scale of Roman politics and its implications for
the individual institutions in Rome, in this nal chapter we will focus on
the connection between politics and society in general. The aim is to
take a broader look at issues raised in the previous chapters and place
them in their proper socio-economic context. The hope is to shed
further light on the relationship between elite and masses, and the
extent to which politics represented an integrative factor connecting
the two.
A central theme in this study of Roman politics has been the scale of
participation. It was argued that it remained very limited, not least when
compared to the size of the city of Rome and with the Roman citizen-
body as whole. Probably no more than a few per cent could attend the
meetings and assemblies, and often the level of attendance would have
been much lower. The implication is therefore that the large majority of
the population never took part in the political process. The Roman
system was, in other words, based on the few rather than the many.
This conclusion is in itself neither new nor surprising. A number of
factors can be adduced, which would have contributed to keeping
attendance down. There were practical difculties posed by the amount
of time taken up by meetings and the lack of remuneration for the lower
classes. Moreover, the assemblies were deprived of any independent
political initiative, and the socio-economic graduation of the inuence
accorded to individual votes left large sections of the urban plebs without
any effective say. On a political level the limited scope of the issues dealt
with in the assemblies would have meant that the proceedings rarely had
much direct relevance to the lives of most citizens.
Even in modern democratic systems mass participation has been
achieved only because the system now works on a representative, rather
than direct, basis, thus reducing people's active involvement to an
absolute minimum. And that despite the fact that politics now plays a
far more pervasive role in society, and virtually every aspect of the lives
of individual citizens may be affected by centrally made policy decisions.
128
Plebs and politics 129
In modern times participation has been further stimulated by the
emergence of permanent political parties, which represent different
interest groups and ideologies and add both clarity and continuity to the
political process. Finally, and perhaps as importantly, the development
of mass media, print as well as electronic, now offers a framework for a
focused public debate on political issues, to which ordinary citizens can
also have regular access.
The differences between modern representative democracies and
republican Rome are too obvious to need much elaboration, and should
merely serve to put our preconceptions about political participation into
perspective. A priori assumptions about the natural character of popular
involvement in politics may be misplaced in a Roman context. They are
essentially projections of modern civic ideologies and practices and
therefore likely to be of limited value in understanding the workings of
Roman politics. The point is that because a system formally entitles all
citizens to a share in the decision-making, we should not automatically
expect to nd mass participation. This is unlikely to be achieved without
effective promotion by the political èstablishment', and that was con-
spicuously absent in the Roman republic.
The main obstacle to the realisation of the `democratic' potential of
the popular institutions remained the particular position of inuence
which the system reserved for the elite. The mere existence of a
permanent body of nobles, who monopolised all political initiative,
experience and authority in the Roman state, would inevitably have
threatened the powers held by the comitia. It is a serious question
whether democratic institutions can function properly within a `mixed'
system of government, since this by its very nature carries an implicit
denial of the people's right to exercise their sovereignty without limita-
tions.
1
For the result of the formal division of powers between the
assembly, the senate and the magistrates would not have been a
`sharing' of inuence, but rather the neutralisation of the popular
element. As the Athenian experience showed, this could full its
purpose only if supported by practical initiatives, facilitating the imple-
mentation of the democratic ideal. In Rome, however, the underlying
rationale behind the political system remained the aristocratic conten-
tion that the voice of the people must be tempered by the moderating
inuence of the senate and the propertied classes. No measures were
therefore ever taken to stimulate large-scale participation, which is
crucial in a system of `direct' democracy based on personal attendance.
1
Cf. the notion of licentia and ìmmoderata libertas', cf. pp. 1012. According to Cicero
the underlying rationale of the Roman political system was to ensure that: `. . . semper
in re publica tenendum est, ne plurimum valeant plurimi', Rep. 2.39.
It also allowed the arcane structures of the popular institutions to
continue for centuries without any substantial changes; no attempts
were made to adapt them to the new circumstances which followed from
Rome's territorial expansion and the growth of her citizen-body.
Thus, the small scale and limited participation may be seen as the
logical consequence of the political set-up in Rome, but it also raises the
question of who made up the minority that did attend the meetings and
assemblies. One of the main problems in dealing with this issue is the
fact that formally these relatively small gatherings represented the entire
Roman people. This was partly a rhetorical convention, but it also
reects a more fundamental split between discourse and reality which
emerged with the expansion of the Roman polity. The problem was that
the notion of the `city-state' continued to dene the Roman self-image
long after the Roman state had outgrown this ideal. As a result the
political identity of the Roman state no longer corresponded in any
meaningful way to the realities of Roman society as it had developed in
the middle and late republic. The last period, in particular, saw a
growing disparity between the ideals and the practice of politics, and a
central feature of this process was the separation of the `people' as a
political concept from the actual masses of Rome. As we have seen, any
crowd convened according to certain procedures was the Roman people,
a fact which affects any attempt to grasp the nature of late republican
politics and makes it very difcult to gauge who was actually present on
these occasions.
The practical and political obstacles to working-class participation
have already been noted, and it follows logically that under normal
circumstances there would be a strong social bias towards the proper-
tied classes. A certain social mix was of course possible, but since the
overall scale of participation was so limited, it becomes evident that
political activity certainly within the ofcial fora cannot have been a
regular or habitual part of the everyday lives of the urban masses. As
Aristotle noted, in politics the distinction between the few and the many
is likely to be one between the rich and the poor. When only a small
section of the population is politically active, the group in question is
unlikely to be socially representative or to include many members of the
lower classes.
It was therefore suggested above that the typical political crowd in
Rome probably represented the rich rather than the poor. While there
was no formal exclusion of the lower classes, the logic of the system
naturally favoured people with time, resources, interest and a certain
level of integration into the world of politics. The order and stability
which traditionally characterised assemblies and meetings may therefore
be explained by the broad consensus which existed between the political
class and the social stratum representing the `people' in the popular
institutions. To the extent that members of the lower classes did turn up
for such occasions, they are likely to have followed established conven-
tions and submitted to the authority of their social superiors.
The late republic saw the end of this consensual system; the popular
institutions now frequently turned against the senate and became an
effective vehicle for anti-senatorial legislation. The process has often
been interpreted as a breakdown of social order and a decline in
traditional bonds of loyalty between the classes. But the notion of the
urban masses released from domination by the elite and rallying to
support their `popular' champions is not really convincing. More likely
we are dealing with new sections of the population being drawn into the
political scene by members of the elite, intent on defying the authority of
the senate. The political upheavals in this period would therefore seem
to reect a breakdown of elite cohesion rather than of social control.
The result of this polarisation was a growing `politicisation' of the
popular institutions.
The ideological importance of contiones made them the natural foci of
`popular' activity, but they might also be useful to their òptimate'
opponents, offering them an opportunity to claim the backing of the
`real people'. But they were effective only if a broadly sympathetic
audience could be counted on, and there are signs that they gradually
turned into more stage-managed `party' events, for which audiences
were mobilised in advance. Unpredictable and anarchic contiones may
still have occurred, but the logic of the institution was to develop into
organised manifestations, which served to demonstrate publicly the
support of the populus Romanus. Viewed in that perspective the form and
signicance of the speeches delivered on these occasions may have to be
reconsidered; informed by long-established rhetorical conventions and
drawing on the traditional libertas ideology, they may have had only a
tenuous connection with the realities of political decision-making in the
late republic.
The collapse of the elite's ability to manage state affairs under the
senate's collective authority also affected the elections, which became
the focus of intense activity by individual nobles. As competition
increased, new methods of campaigning and vote-rigging were intro-
duced, which gradually changed the face of political life in Rome.
However, given the built-in correlation between social standing and
electoral inuence, those most affected probably belonged to the elite.
This interpretation of political practices has repercussions for our
general picture of the relationship between the elite and the masses of
Rome. Thus, the emergence of the `populares' and their success in
passing anti-senatorial bills would suggest that the senate's traditional
political ascendancy had not been based on direct personal control over
the citizen population. Moreover, the recorded events indicate that the
senate was powerless to stop aberrant nobles who deed its authority
and turned directly to the comitia. It appears to have been unable to put
up any serious opposition in the assemblies and had to have recourse to
violence, religious obstruction, tribunician intercession and subsequent
annulment of `popular' laws. The senate's failure to raise effective
counter-crowds is a striking feature of late republican politics. In clashes
between the senate and its opponents the ranks of the former often
appear to have been lled by household slaves and members of the elite
itself. And the difculties faced by the senate whenever it tried to
mobilise popular support are thrown into sharper relief by the limited
scale of politics, which meant that even very small crowds would have
been able to make a difference.
Such difculties were not conned to the optimate elite. For the
populares also, bringing ordinary citizens into politics appears to have
been an up-hill struggle. Despite the size of the urban population they
too had problems mobilising support among the plebs. The early popu-
lares, in particular, had to invest considerable efforts to succeed, and
even suffered the occasional defeat. Moreover, when `popular' leader-
ship was absent from the political scene, the optimate backlashes
suggest that so were the `popular' supporters.
A similar paradox is apparent in the electoral campaigns of this
period. Despite stiff competition between hundreds of candidates all
struggling to raise electoral support by almost any means, the number of
voters turning up for elections seems to have stayed fairly modest. We
have no evidence for any logistical pressures in this period, and the
capacity allowed by the venues and procedures never appears to have
presented a problem. We are therefore left with the impression of a
marked disparity between the efforts put into electioneering, the size of
the urban population and the small scale on which participation never-
theless remained.
The social implications of this analysis of political practices in the late
republic differ substantially from those of other current models. Where
traditional scholarship has stressed the signicance of social control and
personal ties, and the recent `democratic school' pointed to the freedom
and inuence of the plebs, the present reconstruction suggests a society
with relatively limited political and social integration. Late republican
Rome emerges from this inquiry as a place with little contact or
communication between elite and populace, where the world of politics
remained largely separate from the one inhabited by the urban masses.
To many that may seem an unduly negative assessment of Roman
society, but we have to ask whether this model might not be in better
agreement with our general knowledge of late republican Rome than
any of its alternatives. For if we consider the size and structure of Rome
in this period, it would seem more realistic to stress the distance
between the classes than the close integration or the independence of
the plebs.
Rome had by then developed into a very particular kind of society.
Above all it had reached a quite unprecedented scale. Already in the late
republic the population of Rome is likely to have been close to 1 million,
a gure rst reached again by London around 1800. This gure is based
on the numbers of grain recipients recorded in the ancient sources.
According to Suetonius, Div. Jul. 41.3, no fewer than 320,000 beneted
from the Clodian scheme at the time of Caesar;
2
apparently slave-
owners had exploited the system by freeing their slaves in order to make
them eligible for the dole.
3
To this gure must be added women,
children, slaves and foreigners. The proportion of slaves is unknown,
but even a cautious estimate brings the total population to well above
800,000. An often overlooked passage from Livy conrms this esti-
mate.
4
Referring to the late 60s, it mentions that Rome daily required
80,000 modii of grain, suggesting a population of about 1 million.
5
The
sheer size of Rome thus puts the city in a category of its own; we are
dealing with a kind of society which had no direct parallel in the ancient
world and was largely unknown before the industrial revolution in the
late eighteenth century.
6
The social composition of this vast population remains conjectural;
still, some basic features can be gleaned. At the top we nd a small elite
consisting of 300 senators, rising to 600 after the Sullan reform. They
were drawn from the equestrian order, which itself could hardly have
numbered more than a few thousands in the city of Rome. Below this
absolute elite were the upper-income classes made up of men with at
least a modest amount of property, which relieved them of manual
2
See the detailed discussion in Lo Cascio (1997).
3
Dio 39.24.1; Dion. Hal. 4.24.5.
4
Schol. Vratisl. ad Lucan. 1.319.
5
Garnsey (1988) 1912, n.26, estimates an annual minimum consumption per capita of
22.5 modii and an average of 30 modii, cf. Forbes and Foxhall (1982). On this calculation
the 29,200,000 modii, which Rome according to Livy needed annually, would have
provided for 973,333 inhabitants.
6
Arguing on the basis of a comparison with other pre-industrial cities, Storey (1997) has
recently put the gure much lower. His treatment of the recorded gures for grain
distributions and requirement seems supercial, however, and there is a general failure
to appreciate the exceptional position of Rome among pre-industrial cities.
labour. Though technically plebeians,
7
they probably formed part of
Cicero's boni, the pillars of society from whom the senate drew its
primary support.
8
This group did not in any way represent a Roman
equivalent to the modern middle classes; they were distinct from the
political class primarily by their lesser wealth, not by any differences in
economic background, aspirations or general outlook.
9
Outside these
privileged circles we nd the mass of urban citizens who had to work for
a living, many as shopkeepers or craftsmen. At the bottom of this
hierarchy were the day-labourers, who led an even more precarious
existence. The lower classes did not represent a uniform mass; there
were undoubtedly subtle distinctions of status between members of this
group, but they were all set off from the upper echelons of society by
their reliance on their own labour.
The social distance between the top and bottom of society must have
increased in the late republic when enormous wealth poured into the
private coffers of the nobles, the equites and the propertied classes in
general.
10
Huge fortunes were amassed by the political class, which
increasingly outed the traditional codes of modesty and spent large
sums on houses, luxuries and the pursuit of political careers. Their
country and seaside villas mushroomed all over Italy,
11
and in Rome the
7
Cf. Hoffmann (1951) 736; Ku hnert (1991) 1417.
8
Millar (1998) 147 suggested that `bonus' might have a broader social application,
essentially denoting a particular political stance. There is no evidence, however, that
the boni comprised anyone but the well-to-do, cf. Hellegouarc'h (1972) 48493;
Achard (1973). Cicero and others consistently present them as those with property to
defend. His extraordinary attempt in Sest. 97 to give òptimatus' a wider social
denition, including anybody who supported the senate equites, municipales,
negotiatores, even freedmen is itself pure obfuscation and has no bearing on the boni,
who were never subject to similar reinterpretations. Possession of property was integral
to the meaning of `bonus', and the close afliation with the senate associated with this
group stemmed from their common political interests. Thus, Cicero can even complain
in Att. 8.1.3 that not all boni behaved as boni, since some of them supported Caesar.
9
Traditionally historians have seen Roman society as profoundly polarised, consisting
almost exclusively of rich and poor. Thus, Taylor (1949) 5 noted that `The population
seems to have consisted primarily of the well-to-do and the poor, with a very small
middle class', later followed by Brunt (1971a) 383, who saw `no evidence for a middle
class in the city, intervening between them [the equites] and the poor, except for some
rich freedmen'. This picture has increasingly been questioned in recent decades. The
concept of a `middle class' is clearly a red herring in the discussion of ancient Rome,
but the absence of a group with these particular social and ideological characteristics
does not mean that there was no economic stratum situated between the equestrians
and the manual workers and shopkeepers. A priori it seems implausible that the entire
population should have fallen into two social and economic extremes, either rich or
poor. A number of modern studies have accepted that in Rome there must also have
existed a social category between the elite and the working classes; e.g. Christ (1980);
Frier (1980) 42; Millar (1998) 203.
10
See e.g. Shatzman (1975).
11
Cf. D'Arms (1970).
level of domestic luxury grew rapidly in the rst century BC, as
illustrated by their spacious houses and large slave-holdings.
12
By contrast, there are few signs of any improvement in the living
conditions of the lower classes, which generally present a bleak picture
dominated by economic uncertainty, poor housing, food shortages and
debt.
13
The bulk of the urban population would have lived in multi-
storey tenement blocks, which probably represented both a threat to
their health and safety and a burden on their nances.
14
Rents thus
seem to have been high, and the problem of debt, well documented in
the late republic, may therefore have affected not only small traders and
craftsmen (as is often assumed) but also the poorest sections of the
plebs.
15
Simply making a living for themselves would have been a
pressing concern for most members of the lower classes. In order to
provide food and other necessities an income had to be earned on a
daily basis. That might have been a challenge even under normal
circumstances unemployment was not an unknown phenomenon in
ancient Rome. But food crises whether naturally occurring or, in some
cases, provoked deliberately regularly drove up prices far beyond the
means of ordinary plebeians.
16
These crises were unpredictable but
seem to have struck the city with considerable frequency; in the late
republic they are reported on average every four years. Large sections of
12
An example of a noble domus from the late republic has been excavated on the northern
slopes of the Palatine, Carandini (1986); (1988) 35987. Only the basement,
containing the bath and the slave quarters, is preserved, the latter providing sleeping
space for fty slaves. The total number of slaves in the house may of course have been
even higher. Topographical allusions in the ancient sources might indicate that the
domus belonged to Aemilius Scaurus, who was praetor in 56. He was well known for his
extravagance; Pliny mentions that his atrium was supported by four 11-metre tall
columns of `Lucullean marble', NH 36.56.
13
For the living conditions of the lower classes see Yavetz (1958); Bruhns (1981); Scobie
(1986); Drexhage (1989); Whittaker (1993); Prell (1997).
14
Examples of such tenement blocks are known from Ostia, but they may give a false
impression of the quality of housing in late republican Rome, Packer (1967).
Collapsing housing blocks are a commonplace in the late republican sources; together
with the obvious re hazards this represented one of the main risks involved in urban
property investment. A single imperial tenement block from Rome has been
preserved, the Casa di Via Giulio Romano, which offers an important corrective to the
Ostian picture of safe and fairly comfortable accommodation for the lower classes,
Packer (19689). Here the ground oor was occupied by tabernae with living space in
the mezzanines above. The next oor was divided into ats with up to three rooms,
while on the two top(?) oors there were single room units without natural light,
reached by a corridor directly facing the bedrock of the Capitol. These rooms may have
been occupied by entire families, and would presumably have been rented on a short-
term basis, a practice which the legal sources suggest was not uncommon among the
poorest. As Frier has argued (1980) 3940, poor tenants probably enjoyed little legal
protection.
15
Cf. Yavetz (1958); Drexhage (1989) 127ff.; Giovannini (1995).
16
Virlouvet (1985); Garnsey (1988) 198206; Cherry (1993).
the urban population thus lived under the permanent threat of food
shortages.
The evidence suggests that Rome had developed into a large society
with sharp social contrasts and a widening gap between the elite and the
populace. Nevertheless, we often nd it presented as an integrated
community, where the noble lived in harmony next door to the
pauper.
17
Not only does this picture seem somewhat idealised, but the
familiar pattern, known from Pompeii and other small towns, of ne
domus, workshops and lower-class housing all within the same insula,
cannot be traced in the Roman metropolis. Here there seems to have
been a tendency towards social segregation in the urban fabric during
the late republic. The elite concentrated their housing on the top and
slopes of hills, especially the Palatine, where literary and archaeological
evidence suggests a high density of noble houses with no admixture of
plebeian dwellings. Extant information on the houses of the political
elite in the late republic indicates that with very few exceptions they all
lived on the Palatine and the neighbouring Velia.
18
Some nobles,
including Pompey, had houses in the adjoining district of the Carinae,
located at the top of the Via Sacra above the Subura.
19
Caesar, quite
exceptionally, lived in the Subura itself, the popular neighbourhood east
of the Forum traditionally associated with the lower plebs, until he
became Pontifex Maximus and moved to the Domus Publica, Suet. Div.
Jul. 46. Earlier C. Gracchus had transferred his domicile from the
Palatine to the Subura, a move which went completely against the trend
in the late republic, Plut. C. Grac. 12.1. In both cases the choice of
neighbourhood is likely to have reected a deliberate political stance.
These changes are hardly surprising. Rome represented a different
order of magnitude and urban development followed a different logic
there. With the growth of the city social segregation became a desirable
as well as a feasible option for the elite, who had every reason to avoid
the poorer neighbourhoods. Not only would these areas be smelly and
noisy, there would also have been greater risks of re, collapsing build-
ings, riots and violent crime. Logically, therefore, the elite concentrated
their occupation on the attractive hilltops at a safe distance from the
low-lying, frequently ooded areas which were left to the lower classes.
It would be tempting to see in these developments a reection of a
more general fragmentation of the social fabric, a translation of economic
17
Thus Carcopino (1941) 27, who stated that `. . . high and low, patrician and plebeian,
rubbed shoulders everywhere without coming into conict'.
18
Cf. Royo (1987); (1999) 9117.
19
Cf. Varro LL 5.478. Pompey: Cic. Har. resp. 49; Suet. Tib. 15.1; Octavian: Serv. ad
Aen. 8.361. Cicero also lived there until he moved to the Palatine, Plut. Cic. 8.6. Cf.
Ziolkowski (1996); Palombi (1997).
equality into physical distance. The growth of the urban population had
brought an end to the traditional face-to-face society, and the economic
differentiation led to a socially more articulated urban structure. Viewed
in that perspective late republican Rome emerges as an increasingly
divided society, where top and bottom were growing further apart. We
may wonder what level of integration could be maintained under these
conditions.
The issue of clientela has already been raised several times in the
previous discussions of political practice. Suggesting as they did a very
low level of popular participation and limited contact between politi-
cians and the plebs, these investigations also questioned the existence of
the complex networks of formalised links between nobles and plebeians.
Given the size and structure of Roman society in this period, this
conclusion is hardly surprising. Rome had by this time outgrown the
natural limits within which a system of comprehensive, formalised
clientela could be effectively maintained. Rome also experienced con-
siderable immigration and many plebeians therefore had no urban
ancestry or family networks. Since plebs and elite no longer seem to have
inhabited the same social space, late republican Rome would have
offered few opportunities for establishing direct personal contacts across
the class barriers.
It has been suggested that the gap between the top and the bottom of
society might have been bridged by brokers of patronage, middlemen
who used their access to the rich and powerful to provide favours to
those further down the social hierarchy.
20
This model would enable us
to solve the problem of social and physical distance and keep the notion
of clientela as one of the central pillars on which the political ascendancy
of the nobility rested. The existence of brokerage is, however, hypo-
thetical in the Roman republic and may well turn out to be a red
herring.
Brokers of patronage are well attested in later European history,
where they form an integral part of bureaucratic systems as distributors
of favours and resources. As Saller has shown, the phenomenon was also
common during the Roman empire.
21
But here it was intimately linked
to the new political situation where one person had established himself
as the supreme patron and ultimate fount of all public benecia. It thus
worked hand in hand with the centralised imperial bureaucracy, which
had no parallel under the republic either. The imperial evidence also
suggests that brokerage of this kind was located socially far above the
common people and was focused primarily on ofcial honours and
20
E.g. Wallace-Hadrill (1989) 812; Johnson and Dandeker (1989); Flaig (1995) 104.
21
Saller (1982) 745.
positions.
22
Therefore, even in this period brokerage did not represent a
comprehensive system of social exchange connecting on a permanent
basis the lower classes to the top of society; it was a system by which
members of the elite gained access to individuals within the imperial
bureaucracy who were able to inuence the distribution of centrally
controlled favours. As such it cannot be applied to the republic, which
was characterised by a diffusion of political power and the absence of a
centralised bureaucracy. Certainly, it would go too far to present
brokerage as a signicant cohesive factor, able to replace traditional
cliental bonds between the classes.
There can be no doubt that clientela in some form existed in
republican Rome, but it is a big question how many members of the
lower classes would have been included in support networks of this type.
We should not overlook the possibility that the system might have been
inherently biased in favour of the better-off. Given the timocratic nature
of the comitia centuriata and the relegation of the urban masses and the
freedmen to only four tribes, the political benets from wealthy clients
would have been much more valuable. There were also lucrative favours
in the form of gifts and legacies to be gained from clients with prop-
erty.
23
Logically, patronage would have been extended to those who had
something to offer in return, which automatically put the large majority
at a disadvantage. The marginal role of the plebs meant that there was no
political imperative for the elite to cultivate the lower classes.
The incentive to extend patronage to broader sections of the plebs
would have to come, not from political expediency, but from traditional
notions of community. The concept of the free citizenry, representing
the solid foundations of the res publica, still informed the Roman self-
image, and in the Roman mind the classes were still bound together by
common citizenship and a shared heroic past. In principle such notions
might have helped to maintain aristocratic ideals of civic responsibility
and social obligation. But if we look at the evidence we nd few signs of
any paternalism among the nobles, or much concern for the lower
classes in general. Again, it seems that the elite was able to dissociate the
positive concept of the populus Romanus from the mass of ordinary
people who surrounded them.
22
Ibid., 1345.
23
There are indications that wealthy clients were more attractive to the elite. Plautus
refers to the difculties experienced by a poor man trying to nd patronage. Clientela
was not about des but res, he complains; what mattered was `not the client's value as a
man and as a friend, but simply his assets', Men. 5749. In 204 the Lex Cincia had
tried to impose a limit to the `gifts' presented by clients to patrons who had defended
them in court, underlining the fact that these could expect their clients to express their
gratitude in material terms.
The attitude of the upper classes to the common people seems
largely to have been one of contempt. Like many other aristocracies
throughout history, the Roman elite viewed the lower classes as morally
and intellectually inferior. Thus, our sources from the late republic
invariably promote the familiar stereotype of the poor as depraved and
untrustworthy, poverty routinely being associated with crime and
subversiveness.
24
Historically these attitudes may not be exceptional, but in Rome they
seem to have been taken further than in most other oligarchic societies.
The automatic equation of penury with moral inferiority was so in-
grained that in Rome ègens', poor/needy, became a common term of
abuse, which somewhat paradoxically could be used purely in its
moral sense as a term of abuse against other well-to-do men.
25
Fol-
lowing the same logic `locuples', rich, also took on a broader meaning
and was used in aristocratic circles as a term of praise.
26
The basis for
this elevation of wealth as a source of personal virtue lay in the
aristocratic belief that only the rich man had freedom of choice and thus
was able to act according to moral principles. The poor man was under
the law of necessity, which imposed its own morality or rather lack of
morality. As Publilius Syrus stated, `necessity makes the poor man a
liar'.
27
It followed that a poor man could have no honour. By inclination
and circumstances he was untrustworthy, àudax' and `perdiosus', Cic.
Cluent. 70. Poverty was thus akin to servitude, the ultimate state of
necessity, where no morality was possible and neither integrity nor
honesty could be expected. Logically the testimony of slaves was
accepted in court only if given under torture. Similarly, witness state-
ments provided by the poor were also considered suspect.
28
Material necessity forced people to perform tasks which the elite
considered to be demeaning for a man of honour. Most obviously it
24
E.g. in the common connection of ègens' and àudax', `facinerosus', `perditus',
ìmperitus', ìmprobus', cf. Cic. Mil. 36; Dom. 45; Att. 9.7.5; Agr. 1.22. Cf. Wood
(1988) 96; Prell (1997) 21719.
25
In Verr. 2.94 Pacilius is called ègens et levis'; Sex. Cloelius ègentissimus' and
`facinerossimus', Dom. 25, and a tribune of the plebs `hominem nequam atque
egentem', Att. 1.19.5. In Flacc. 52 Maeandrius from Tralles is described as `homini
egenti sordido, sine honore, sine existumatione, sine censu'. Likewise in Comm. pet. 8
Catiline and Antonius are described as: `competitores ambo a pueritia sicarii, ambo
libidinosi, ambo egentes'. Cf. Ku hnert (1989) 4378; Prell (1997) 449.
26
Thus, Cicero, Att. 5.20.4, describes T. Gavius Caepio as a `locuples et splendidus
homo', and Q. Cicero makes the same connection in Comm. pet. 53, mentioning `viri
boni ac locupletes', cf. Phil. 13.23.
27
Sent. N31: `Necessitas egentem mendacem facit.' Cf. Sall. Cat. 37,8: `homines egentis,
malis moribus'.
28
In his speech for Flaccus, 5 frg. Mediol., Cicero thus refers to ègentissimos testis',
suggesting they were patently unreliable.
compelled people to sell their labour in return for wages, which in the
eyes of the elite effectively reduced them to the level of slavery.
29
But
similar opprobrium was attached also to retail and small-scale trading
which was considered sordid, because it was unproductive and therefore
involved deceit. The `poor' were, in other words, not simply the
completely destitute, but logically included all those who lived under the
law of necessity, thus comprising also ordinary craftsmen and tabernarii.
This point is made explicitly in Cicero's mention of Lentulus' `leno',
pimp, who ìs making the rounds of the shops, hoping to buy the
support of the poor and the ignorant'.
30
The `poor' were simply dened
as those without (landed) property, which could give them freedom and
preserve their dignity and honour. This category thus included the large
majority of the urban plebs, who could therefore command no respect or
trust, either individually or as a body.
The elite's contempt for the masses was often tinged with an element
of fear. The poor were typically perceived as the natural opponents of
the rich, even their enemies. In his speech to the senate on the rogatio
Servilia, Agr. 1.22, Cicero invoked the spectre of an army of the poor
and wretched mobilised against the established order, and in De domo
sua one of Clodius' men, Sergius, is accused of: `plotting a sudden onset
upon the consuls, the senate, and the property and fortunes of the rich,
in support of the destitute and ignorant'.
31
Similarly, in Mil. 95 Cicero
warns his imaginary jury that the `plebs et inma multitudo' under
Clodius' leadership threatened `fortunis vestris'. The fundamental
dichotomy between the elite and the plebs is made quite explicit in
Pro Plancio, where Cicero invoked the dreaded scenario of `the poor
being armed against the rich, perditi against boni, slaves against their
29
In antiquity no clear distinction was drawn between selling your labour and selling
yourself. Wage labour was therefore logically perceived as a form of short-term slavery.
Thus, most famously Cic. Off. 1.1501, following Arist. Pol. 1337
b
421, cf. Joshel
(1992) 67. Treggiari (1980) has argued that legally manual work was not placed on the
level of servitude. Still, that does not imply that ideologically the elite did not make this
connection, linked as the two seemed to be by a common lack of freedom. Slaves and
workers were all at the mercy of external powers, bondage and material need,
respectively. A slave was, according to Seneca, Ben. 3.22.1, a `perpetuus mercennarius',
cf. Mo ller (1993).
30
Cat. 4.17: `concursare circum tabernas, pretio sperare sollecitari posse animus
egentium atque imperitorum . . .' Similarly in Flacc. 18 Cicero mentions òpices' and
`tabernarios' in the same breath as ìllam omnem faecem civitatum'. Cicero's
celebration of social peace in Cat. 4.1617, which includes a rather positive image of
the tabernarii, is most untypical. Cf. Cicero's reference to tabernarii as ìnopes atque
imperiti', Dom. 13, and invective against `homines paene operarios', Rosc. Am. 120.
31
Dom. 13: `. . . in consules, in senatum, in bona fortunasque locupletium per causam
inopum atque imperitorum repentinos impetus comparares . . .'
masters'.
32
The lower classes are here clearly perceived as a latent threat
to the elite. In this conict Cicero identied the rich as his army,
remarking to his friend Atticus that: `this, as you know, is my army the
well-to-do'.
33
The sources reveal little trace of aristocratic paternalism or any sense
of community with the lower classes. Again, the profound respect for a
notional populus Romanus, professed by all Roman politicians
34
, went
hand in hand with a disdain for the actual people, highlighting not only
the ambiguous nature of this concept in Roman politics but also the
distance between the elite and the populace. Given their attitude to the
lower classes, it is difcult to imagine Roman nobles seeking a following
of common people for their own sake. Logically we might therefore
draw the conclusion that the large majority of them were most likely left
to fend for themselves in their local neighbourhoods, away from the
world of the elite and Roman politics.
Viewed in that perspective the acute crisis which confronted the
senate in the late second century may appear in a different light. The
vulnerability of the ruling order was highlighted for the rst time by its
`popular' opponents, who used the plebeian vote to carry anti-senatorial
legislation. The senate found itself without any effective political means
of responding to this challenge and had to have recourse to obstruction,
annulment and a spurious right to declare states of emergency in order
to justify violent interventions. The fundamental problem seems to be
that the elite had allowed itself to be increasingly separated from the
masses, perhaps believing that its political monopoly was complete and
unchallengeable. Eventually, however, the real threat came from within,
and at that point it seems the senate had no effective way of reaching the
plebs. Certainly the history of the late republic would suggest that it was
no longer possible to mobilise them on behalf of the ruling class.
Their `popular' opponents may not have been much better connected,
but they gradually developed new strategies to overcome the distance
between plebs and politicians. Their rst attempts were understandably
tentative and not altogether effective. Given the scale and fragmentation
of the urban fabric, the approach had to be decentralised and concen-
trate on the local neighbourhoods, the vici. They represented the focus
of the daily lives of the urban masses, and the local structures which
already existed there in the form of associations and festivals could be
32
Planc. 86: `. . . egentes in locupletes, perditi in bonos, servi in dominos armabantur'.
The same vision of society is spelt out in Sall. Cat. 37.3: `Nam semper in civitate
quibus opes nullae sunt bonis invident, malos extollunt, vetera odere, nova exoptant,
odio suarum rerum mutari omnia student . . .'
33
Att. 1.19.4: ìs enim est noster exercitus hominum, ut tute scis, locupletium'.
34
Aptly described as `rhetorical genuexions to the Populus Romanus', Wood (1988) 96.
used to reach the lower-class inhabitants in the area. Their importance
for popular mobilisation is underlined by the senatorial ban on collegia
and neighbourhood festivals, which must be seen as a deliberate attempt
to ensure that the people's traditional absence from the political scene
was maintained. The ban was overruled by Clodius who further rened
the system by creating a proper command structure which involved both
the local ofcials and his own personal assistants.
35
The centrality of the
vici to Clodius' success is illustrated by the fact that his four main laws
were passed immediately after his celebration of the Compitalicia in
deance of the senate's ban; apparently he had used the festival to gain
popularity and prepare the mobilisation of the plebs two days later.
36
His local network of contacts enabled Clodius not only to dominate
the popular assemblies, but also to overcome the inherent short-
termism of Roman politics, which posed yet another obstacle to an
effective mobilisation and `politicisation' of the lower classes. The
fact that Roman politics was structured largely around one-man political
`machines', created ad hoc for specic electoral purposes, made it
difcult to establish comprehensive networks or maintain lasting con-
tacts with broad sections of the urban population. A Roman noble
seeking ofce would remind his friends and acquaintances of past
favours, approach potential new supporters within the elite and promise
them future support; he would also arrange public appearances aimed at
improving his public image and raising his prole. As the Commentar-
iolum suggests, campaigning was about reactivating existing connections
and creating new ones for the sole purpose of meeting the immediate
challenge which lay ahead, that is the forthcoming election. This had
two important consequences for the feasibility of popular mobilisation.
35
Fraschetti (1990) 244, sees Clodius' mobilisation as a centralised operation, focused on
the Forum. But the passages of Cicero used to support the argument may be more
ambiguous. Thus, e.g. Dom. 54, where Cicero asks Clodius whether he was not
preparing for violent action, when ìn tribunali Aurelio conscribebas palam non modo
liberos sed etiam servos ex omnibus vicis concitatos'. The point here is that Clodius
had brought his followers from the vici to the Forum in order to register them openly in
what was clearly a deliberate provocation against the senate and the boni, whose space
he and his followers invaded. Similarly, Clodius' demonstration in 56, when his force
èx omnibus vicis collecta' burst into the Ludi Megalenses during their celebration on
the Palatine, Cic. Har. resp. 22. Again people from the vici were mobilised locally and
then thrown into action elsewhere in the city.
36
The Compitalicia had been celebrated on 31 December 59, while Clodius' leges
frumentaria, de obnuntiatione, de collegiis and de censoria notione were all passed on 2
January 58. As Fraschetti (1990) 21017 has pointed out, the Compitalicia were foci
of popular unrest and thus useful for political mobilisation. Earlier Manilius had used
the same strategy; passing his controversial bill on the tribal inscription of freedmen
during the celebration of the Compitalicia, Asc. 45C; Dio 36.42. According to Sallust,
Cat. 18.5, the rst Catilinarian conspiracy had also been planned to erupt on the rst of
January 65.
On the one hand, the personalised form of electioneering automatically
limited the scope of the campaigns to the nobles and the upper classes;
members of the plebs were apparently approached only indirectly
through their local leaders. On the other hand, it implied a rather short
time horizon and a decentralisation of the campaigning efforts of the
political class as a whole. Since the focus was on specic electoral
contests, many contacts may have become redundant after an election
had been successfully fought, and would lie dormant until the next
attempt was made to move up the career ladder. Moreover, the fact that
each candidate worked alone at least in principle meant that, despite
the enormous resources and energy invested in electoral campaigning in
the late republic, the overall impact on the relationship between plebs
and politicians was bound to be limited. For the implication was that
most candidates would have approached the same narrow circles lying
within reach of their campaigns. The fact that there were no permanent
`party-machines' working between elections also made it difcult to
maintain or transfer any contacts which might have been established
with representatives of the broad population during a campaign. It is a
paradox of Roman politics that the personal networks which were pieced
together with such painstaking effort during the campaigns may have
functioned only at those particular moments in a politician's career
when he was actively seeking ofce. There seems to have been a certain
ad hoc aspect to Roman politics which made large-scale campaigning
almost impossible and thus prevented its leaders from wielding contin-
uous inuence among the electorate.
The Roman politicians themselves recognised the problem and made
various attempts to tackle it during the late republic in response to the
ever-increasing competition. Thus, a characteristic feature of this period
was the strong emphasis which leading politicians placed on their
personal tribes, whose most prominent members were courted through
regular donations of sportulae, theatre tickets and so on. Cicero in his
forensic speeches presents this practice as a time-honoured custom, but
as always when Cicero uses that line of defence we have reason to be
suspicious. In fact there are few traces of this practice before the late
second century, and it is therefore likely to be a new departure in late
republican politics.
37
For by cultivating close links with their tribus it was
possible for politicians to maintain some electoral inuence beyond the
actual year in which they ran for ofce. It gave them vital bargaining
power in dealings with other politicians, as clearly indicated by Cicero,
who measured his own tribal inuence against that of Lucceius; both
37
Divisores are for example not documented before this time, cf. Lintott (1990) 78.
politicians had promised their tribes to a third party but only Cicero
could deliver, Att. 2.1.9.
An even more blatant attempt to overcome the inherent short-
termism and decentralisation of Roman networking was marked by the
emergence of sodalitates, informal associations of politicians who worked
together in order to further their individual careers. Eventually the
sodalitates became synonymous with organised bribery and were banned
in 56,
38
but this aspect should not distract us from the fact that these
associations represented a radical innovation in Roman politics, which
challenged fundamental principles of the aristocratic republic. Most
crucially these organisations remained active between elections and
included among their members both politicians currently running for
ofce and prospective candidates who intended to do so in the future.
This new structure made it possible to maintain networks over longer
periods and thus provided an element of continuity which had pre-
viously been absent from Roman campaigning. This allowed politicians
to pool their inuence, which could be shared between several candi-
dates taking part in the scheme or traded with outsiders. Thus, it
appears that Cicero had been promised support from a number of
sodalitates in the run-up to his consular election, Comm. pet. 19. If these
associations had been allowed to develop further, they would almost
certainly have transformed the nature of Roman politics. The senate,
however, insisting on maintaining the traditionally low level of political
organisation, came down heavily on the sodalitates, which went under-
ground and specialised entirely in illicit bribery. The result was that the
organisations needed to cultivate and maintain comprehensive political
networks never came into existence in republican Rome. Popular mobi-
lisation was therefore destined to be of limited extent and duration.
Despite the manifest interest of the political class in raising their
occasional support, the decentralised nature of Roman politics and
the entrenched opposition of the senate prevented the creation of
general networks which might have been able to bridge the widening gap
between plebs and politicians in the late republic.
The conclusion offered by this analysis must be that despite the
polarised climate of the late republic the people of Rome never
became fully integrated into the political process. There was no formal
exclusion but their participation was discouraged by a variety of institu-
tional and practical factors, ultimately all rooted in the fundamentally
aristocratic structure of Roman society. This assessment runs directly
counter to the `democratic' model espoused by a number of modern
38
scholars, and it may therefore be useful briey to set out the main points
of disagreement.
The `democratic' interpretation has sprung from a growing realisation
of the concrete nature of Roman politics and its rootedness in time and
place. Thus, by focusing on the physical interaction between politicians
and people, the practical procedures and the settings, the essentially
public nature of the political process has become apparent. Roman
politics took place in full public view, and the public clearly had a
signicant role to play in the proceedings.
This observation is itself uncontroversial, but the emphasis on the
practice of politics then leads to the conclusion that the system had
features which could reasonably be described as `democratic'. It is this
nal syllogism which is problematic, for it neatly side-steps the crucial
questions of who this `public' was and how much inuence the actual
masses had on the running of the Roman state.
It is a paradoxical consequence of the `democratic' interpretation that
despite its insistence on broadening our understanding of Roman
politics by accentuating physical and practical aspects, the result has in
fact been a narrowing of the social perspective applied to the issue.
What we nd is an almost decontextualised vision of Roman politics.
The image of a `democratic' Rome has very much been achieved by
isolating political practices from their demographic, social and economic
context. We are presented with a scenario of a public actively engaged in
the political process, but the composition of this public is never con-
sidered, probably because in principle all Roman citizens were entitled
to join it. The Roman `democracy' is thus founded on two themselves
indisputable historical facts: the existence of a politically signicant
`public' and the open access of all citizens to participate in this `public'.
These two facts do not, however, add up to a Roman `democracy'. One
crucial factor has been left out of the equation, which is the distinction
between formal rights and their practical realisation. The model there-
fore sits uneasily between, on the one hand, a very practical hands-on
approach to politics and, on the other hand, an idealistic, almost naive
view of the relationship between constitutional principle and reality.
The present study has tried to explore the possible discrepancies
between the two, departing from the simple question of scale, which
immediately makes it clear that politics was an activity reserved for a
very small minority in Rome. We are therefore forced to distinguish
between the political `public' and the `people'. Simply assuming that the
populus and masses were the same merely perpetuates a particular
ideological construction of the `people', rst conceived by the Roman
ruling class.
It would seem a truism that political rights do not exist in a vacuum,
but are embedded in social and economic structures which determine
the extent to which they can be realised in practice. If we accept that
Rome was a society with stark social contrasts between a small,
immensely rich elite, controlling all political ofces and religious
authority, and a vast, impoverished under-class, the structural con-
straints this inequality imposed on popular participation become too
obvious to ignore.
Moreover, as soon as we place politics in its proper social context it
also becomes clear that the `people' had interests distinct from those of
the ruling class. We cannot simply assume that what constituted political
issues to the elite did so to the mass of the urban population also; the
sources, as we know, are conspicuously partial and one-sided on this
point. Given the distance between politicians and the masses, the idea of
a single unied political agenda is implausible. The interests of the
masses would by necessity have been different, and there are many
indications that they were focused on pressing material concerns, that is
food supply, rents and housing, debts and so on. Their ability to
promote these interests is crucial to any assessment of the level of
`democracy' at Rome. If political rights only involved responding to
matters relating to the internal affairs of the elite, then they were
obviously of little practical value. Just as the proof of the pudding is in
the eating, so a democratic system must reveal itself in the opportunities
it offers the masses for actively furthering their own interests.
The obvious test case is the grain provision for the city, and looking at
the history of this issue the conclusion seems inescapable that the people
had little power to set the agenda or enforce its implementation. For
despite the enormous wealth which poured into the treasury in the
second century and enabled the senate to suspend the tributum for the
propertied classes in 167, nothing was done to alleviate the plight of
the lower classes. That did not happen until 122 when the tribune
C. Gracchus rst introduced subsidised grain for sections of the urban
plebs against the will of the senate. His provisions were reduced by a
Lex Octavia and completely abandoned by Sulla in 81, Sall. Hist.
1.55.11, only to be reintroduced on a modest scale in 73 under the
Lex Terentia Cassia. In 63/62 when the senate tried to defuse the crisis
after the conspiracy of Catiline, Cato again expanded the scheme, Plut.
Caes. 8.4; Cat. Min. 26.1. Free grain was not made available until 58,
when the Lex Clodia was nally passed; nevertheless, the dole was still
not sufcient to support an entire family.
39
39
Garnsey (1988) 21114.
Clearly this summary does not suggest a functioning `democracy' at
work; only late and with great reluctance did the senate agree to grant
the people a modest share in the prosperity of the empire. Rather, it
brings out the inherent limitations to the people's power; the system did
not allow them to operate autonomously or formulate their own policies.
Popular attempts to raise the grain issue were made, but that did not
happen in the ofcial fora and institutions. Instead they took place in
open, uncontrolled public spaces; like most under-classes throughout
history the Roman plebs took to the streets whenever it wanted to assert
its interests.
40
Public demonstrations, often caused by food shortages,
were a common occurrence in Rome, which in itself is a serious
indictment of the ability of the assemblies to serve as outlets for popular
grievances and concerns. There is, moreover, no indication that these
measures were effective in promoting their interests. Yavetz noted:
`Democracy did not exist in Rome, but popular pressure did';
41
still, the
course of events would suggest that the republican elite was well able to
withstand it. Thus, the senate's only major concession, the Lex Porcia,
was made in response to a much more serious threat than a common
food riot, that is Catiline's attempt to overthrow the old order.
The voice of the plebs was muted by the fact that it had no representa-
tives, elected on a political platform and obliged to serve their constitu-
ency. As a rule, popular concerns only entered the ofcial agenda when
`popular' politicians happened to adopt their cause. The `power of the
people' was, in other words, realised only through internal dissent
within the elite, which as a whole continued to monopolise political
initiative. Aberrant nobles might promote the interests of the masses,
but even then the elite often succeeded in nullifying or seriously
modifying the measures.
The public nature of Roman politics represented no modication of
the aristocratic system, either ideologically or practically. The internal
structure of the elite, which extended beyond the active ofce-holding
class, automatically opened the curia towards a wider public. Moreover,
the position of the political class was itself justied in terms of a libertas
ideology, which further strengthened the public aspect of politics. But
the public who were recognised as legitimate partners in government
were not the generally despised urban masses but a broader section
of the propertied classes, the boni.
42
They represented the natural
political class and it was from this social stratum that the elite sought
40
See catalogue of documented instances in Vanderbroeck (1987) 22067.
41
Yavetz (1969) 39.
42
Thus, in Sest. 140 Cicero identied the real political authority with the boni, rather than
the ìmperita' and `concitata multitudo'.
political legitimisation. Nothing illustrates the political mind-set better
than the public justication of the so-called Senatus Consultum ultimum.
On a number of occasions Cicero stressed that the moral authority
behind the SCU was derived from the consent of the equites and the
boni,
43
thus suggesting that the political order was based on a general
consensus among the propertied classes. When the senate, the equites
and the boni acted in unison, any established convention could be
overruled.
43
In Rab. Perd. 2, Cicero claimed that the prosecution of Rabirius for his part in the
killing of Saturninus was an attack on àuctoritas senatus . . . consulare imperium . . .
consensio bonorum'. The same line of defence was used to justify his own execution of
the Catilinarians, Dom. 94: `. . . quod ex auctoritate senatus consensu bonorum
omnium pro salute patriae gessissem . . .' Thus, behind him in 63 had been: `tanto
studio senatus, consensu incredibili bonorum omnium', Sest. 36. The later attacks were
therefore not merely targeted at himself, but: `petita est auctoritas vestra . . . consensio
bonorum omnium . . .', Har. resp. 45.
149
Appendix
THE ` LEX LICINIA DE SODALITATIBUS'
The Lex Licinia is central to the study of political participation in the late
republic. It has attracted considerable debate and a range of different interpreta-
tions have been brought forward. It has been seen as a senatorial attempt to
quash the political clubs of Clodius, an attempt to curb electoral malpractice
among the elite, and as a combination of the two, targeting both ambitus and
political violence in general.
1
The senate rst issued a decree on these matters on the tenth of February 57
a week after Clodius' supporters had given Pompey a serious heckling at the
trial of Milo. According to Cicero the Senatus Consultum obliged the consuls: ùt
sodalitates decuriatique discederent lexque de iis ferretur', Q. Fr. 2.3.5. And a
year later Crassus passed his Lex Licinia de sodalitatibus. There are several
ancient references to the selection of judges prescribed in the law, Cic. Pis. 94;
Phil. 1.20; Asc. 21C. But the most important source on the subject is Cicero's
speech for Plancius.
Plancius was accused of bribery according to the Lex Licinia: `reus de
sodaliciis petitus est lege Licinia, Schol. Bob. 152 (St)'. It is apparent that the
law, for the rst time, dened a crimen sodalicium: `quos tu si sodalis vocas,
ofciosam amicitiam nomine inquinas criminoso', 46, and Plancius was formally
prosecuted: `nomine legis Liciniae quae est de sodaliciis', 36. Likewise, in Cael.
16 Cicero refers to the Lex Licinia as being: `de ambitu et de criminibus istis
sodalium ac sequestrium'.
The character of the sodalitates in question is a matter of contention. Cicero,
Planc. 37, gives this description: `cuiuscumque tribus largitor esset, et per hanc
consensionem quae magis honeste quam vere sodalitas nominaretur quam
quisque tribum turpi largitione corrumperet . . .' The scholiast further expands.
`. . . in eos candidatos qui sibi conciliassent [sodales] ea potissimum de causa, ut
per illos pecuniam tribulibus dispertirent ac sibi mutuo eadem suffragationis
emptae praesidia communicarent', Schol. Bob. 152 (St). These passages clearly
suggest that we are dealing with elite associations which organised electoral
1
The second view goes back to Mommsen (1843) 412, 47, and Waltzing (18951900)
I, 97, 112. Later Kornemann (1900) 3901 suggested the rst interpretation, followed
by De Robertis (1938) 1009; Treggiari (1969) 1767; Flambard (1977) and (1981);
Ausbu ttel (1982) 912. Linderski (1961) formulated a compromise which was adopted
by e.g. Shackleton Bailey (1981) 178; Lintott (1968) 219 and (1990) 9; Nadig (1997)
5967. Another attempt to reconcile the two positions was made by Venturini (1984).
150 Appendix. The Lex Licinia de sodalitatibus
bribery. In the same direction also points Asconius' reference to the bribery case
against Milo, 39C: `[Milo] postulatus autem erat et de sodaliciis et de ambitu',
suggesting a similar connection between sodalitates and bribery. Another
earlier reference to these associations comes from Comm. pet. 19, where Cicero
is told by his brother that: `quattuor sodalitates hominum ad ambitionem
gratiosissimorum tibi obligasti . . .' At this stage the sodalitates merely appear as
elite associations organising mutual electoral support without the direct involve-
ment of bribery.
2
Later, however, this aspect became so common that member-
ship of a sodalitas was legally dened as a `crimen' and sufcient to justify public
prosecution for bribery. Thus, Cicero argues that Plancius: `habuisse in petitione
multos cupidos sui gratiosos, quos tu si sodales vocas, ofciosam amicitiam
nomine inquinas criminoso' Planc. 46. The implication is that by then the
sodalitates were so widely associated with electoral bribery that the words had
become largely synonymous. The Lex Licinia therefore emerges unequivocally
from these sources as an ambitus law, targeting electoral malpractice.
3
The case for the Lex Licinia as a measure against Clodius' clubs remains
weak. These particular associations are never described as sodalitates/sodalicia,
but only as collegia, e.g. Cic. Sest. 55, or as travesties of collegia, `simulatione
collegiorum', Post red. sen. 33.
The decuriati, mentioned in both the SC and Pro Plancio, have been seen as a
reference to Clodius' men who were also described in a similar way in Cic. Dom.
13: `decuriatos ac discriptos haberes exercitus perditorum', and Sest. 34: `cum
vicatim homines conscriberentur, decuriarentur'.
4
Here, however, the context is
entirely different from the one obtaining in Plancius' ambitus trial, since Clodius
is accused of having organised street gangs, not electoral bribery. `Decuriare'
was a common term to describe the division of groups of people into smaller
units. Later in Phil. 7.18, Cicero for example refers to the `decuriatio' of
ìmprobi', and there is plenty of epigraphic evidence for collegia and other
associations being subdivided into decuriae. It cannot therefore be taken as a
technical term exclusively referring to Clodius' clubs.
The `decuriatio' of which Plancius stood accused was clearly different from
Clodius' mobilisation of strong-arm men. It was part of a bribery operation
which Cicero describes in detail: `Decuriatio tribulium, discriptio populi, suf-
fragia largitione devincta severitatem senatus et bonorum omnium odium ac
dolorem excitarunt', 45. Later Cicero asked the prosecutor to prove that
Plancius `decuriasse . . . conscripsisse, sequestrem fuisse, pronuntiasse,
divisisse', ibid., and again in 47 that he `. . . sequestrem fuisse, largitum esse,
conscripsisse, tribules decuriavisse'. `Decuriatio' here emerges as a specic
element in a complex bribery scheme, which involved enlisting tribules,
2
Cf. Comm. pet. 18, describing members of sodalitates: `nam per hos annos homines
ambitiosi vehementer omni studio atque opera elaborarunt ut possent a tribulibus suis
ea quae peterent impetrare', the emphasis on `per hos annos' suggesting the relative
novelty of the phenomenon.
3
Cf. Planc. 36; Cael. 16; the fragment of Pro Vatin. IV 3 p. 285 Mu ller; Schol. Bob. 160
(St): Ìam de sodaliciis causam dixerat P. Vatinius eodem defendente M. Cicerone'.
Vatinius stood accused of bribery.
4
Cf. Post red. Quir. 13: `cum homines in tribunali Aurelio palam conscribi centuriarique
vidissem'.
Appendix. The Lex Licinia de sodalitatibus 151
organising them into smaller groups, depositing funds, promising bribes and
distributing them among those who had been enlisted. In the context of electoral
bribery `decuriatio' thus seems to have had a clearly dened meaning, which
could be targeted by legislation.
This use of `decuriatio' had nothing to do with the political street violence
orchestrated by Clodius. Moreover, the `decuriatio' associated with these gangs
may not have been organised along tribal lines in the same way as Plancius'
allegedly had been linked as it was to the structure of the tribal assembly which
elected the aediles. The mention of the tribus Collina in Cic. Mil. 25 is most likely
a reference to the topographical area in which Clodius on this occasion
mobilised his followers.
5
There are, in sum, no real points of resemblance
between the clubs of Clodius and the `decuriati' described in Pro Plancio. While
the former were gangs used for political intimidation and demonstrations, the
latter were tribules enlisted in bribery schemes organised by elite associations.
`Decuriati' are associated with the Lex Licinia only in the context of bribery and
closely connected with the use of sequestres and divisores, cf. Cael. 16: `criminibus
istis sodalium ac sequestrium'. The coincidence in terminology is of no real
import since `decuriatio' was a common method of organising groups of any
kind.
6
The political circumstances surrounding the passing of the senate's decree `de
sodalitatibus' have been invoked as an argument for a `Clodian' interpretation of
the law.
7
That remains highly speculative. Cicero himself does not draw any
connection between Clodius' attack on Pompey and the SC de sodalitatibus, both
mentioned in the same long letter to Quintus. A nal argument for this version
has been drawn from the clause in the SC, which prescribed that the penalty
should be the same as for vis, violence. That, however, does not necessarily
imply that the Lex Licinia itself dealt with vis. The clause merely put the offence
on the same level of seriousness as vis.
In Planc. 36 the Lex Licinia features as one among other laws which dealt with
ambitus, and it is difcult to see how Clodian street gangs could be covered by
the same piece of legislation. Political violence was not a common feature of
electoral campaigning; most often it occurred in the context of legislative
assemblies and court cases. The evidence therefore suggests that the Lex Licinia
was concerned with a particular kind of electoral bribery, which was organised
on a collective basis by groups of nobles who were themselves running for ofce
or intended to do so in the near future. The idea that Clodius' bands might also
have been affected by the law has no rm support in the ancient evidence.
5
`Clodius `convocat tribus, se interponebat, Collinam novam [novo?] dilectu perditissi-
morum civium conscribebat', cf. Linderski (1961) 11415.
6
See e.g. Diz. Epig. 2,2 (1910) s.v. `decuria', 150413.
7
E.g. Treggiari (1969) 175.
Bibliography
Achard, G. (1973) `L'emploi de boni, boni viri, boni cives et de leurs formes
superlatives dans l'action politique de Ciceron', Etudes Classiques 4:
20721.
Aigner, H. (1978) `Gab es im republikanischen Rom Wahlbestechungen fu r
Proletarier?', Gymnasium 85: 22838.
Astin, A. (1967) Scipio Aemilianus (Oxford).
Ausbu ttel, F. M. (1982) Untersuchungen zu den Vereinen im Westen des romischen
Reiches (FAS 11: Kallmu nz).
Badian, E. (1972) `Tiberius Gracchus and the beginning of the Roman revolu-
tion', ANRW 1.1: 668731.
(1989) `The scribae of the Roman republic', Klio 71: 582603.
Balsdon, J. P. V. D. (1969a) Life and Leisure in Ancient Rome (London).
(1969b) `Panem and circensem', in Homages Renard (Brussels) II: 5760.
Beard, M. and M. H. Crawford (1985) Rome in the Late Republic (London).
Bell, A. J. E. (1997) `Cicero and the spectacle of power', JRS 87: 122.
Benner, H. (1987) Die Politik des P. Clodius Pulcher:Undersuchungen zur Denatur-
ierung des Clientelwesens in den ausgehenden romischen Republik (Historia
Einzelschriften 50: Stuttgart).
Berg, B. (1997) `Cicero's Palatine home and Clodius' shrine of Liberty:
alternative emblems of the republic in Cicero's De domo sua', in Studies in
Latin Literature and Roman History VIII, ed. C. Deroux, (Collection
Latomus 239: Brussels) 12243.
Bernert, E. (1936) `Nomenclator', RE 17,1: 81720.
Bernstein, A. H. (1978) Tiberius Gracchus: Tradition and Apostacy (London).
Best, E. E. (1974) `Literacy and Roman voting', Historia 23: 42838.
Bleicken, J. (1974) Staatliche Ordnung und Freiheit in der romischen Republik (FAS
6: Kallmu nz).
(1975) Lex Publica: Gesetz und Recht in der romischen Republik (BerlinNew
York).
Bonnefond-Coudry, M. (1989) Le senat de la republique romaine de la guerre
d'Hannibal a Auguste (BEFAR: 273: ParisRome).
Briscoe, J. (1992) `Political groupings in the middle republic: a restatement', in
Studies in Latin Literature and Roman History VI, ed. C. Deroux (Collection
Latomus 217: Brussels) 7083.
Broughton, T. R. S. (19501) The Magistrates of the Roman Republic, 2 vols.
(New York).
152
Bibliography 153
Bruhns, H. (1981) Àrmut und Gesellschaft in Rom', in Vom Elend der Handar-
beit, ed. H. Mommsen and W. Schulze (Stuttgart) 2749.
Brunt, P. A. (1966) `The Roman mob', Past and Present 35: 235.
(1971a) Italian Manpower 225 BCAD 14 (Oxford).
(1971b) Social Conicts in the Roman Republic (London).
(1980) `Free labour and public works at Rome', JRS 70: 81100.
(1988) The Fall of the Roman Republic and Related Essays (Oxford).
Burckhardt, L. A. (1988) Politische Strategien der Optimaten in der spaten
romischen Republik (Historia Einzelschriften 57: Stuttgart).
(1990) `The political elite of the Roman Republic: comments on recent
discussion of the concepts nobilitas and homo novus', Historia 39: 7799.
Carafa, P. (1998) Il Comizio di Roma dalle origine all eta di Augusto (Rome).
Carandini, A. (1986) `Domus e insulae sulla pendice settentrionale del Palatino',
Bull. Com. Arch. 91: 26278.
(1988) Schiavi in Italia: gli strumenti pensanti dei Romani fra tarda Repubblica e
medio Impero (Rome).
Carcopino, J. (1941) Daily Life in Ancient Rome (London).
Carettoni, G., and A. M. Colini, L. Cozza, G. Gatti (1963) La pianta marmorea
di Roma antica (Rome).
Cerutti, S. M. (1998) `P. Clodius and the stairs of the Temple of Castor',
Latomus 57: 292305.
Cherry, D. (1993) `Hunger at Rome in the late Republic', EMC 37: 43350.
Christ, K. (1980) `Grundfragen der ro mischen Sozialstruktur', in Studien zur
antiken Sozialgeschichte: Festschrift Friedrich Vittinghoff, ed. W. Eck,
H. Galsterer, H. Wolff (Cologne) 199228.
Cloud, D. (1989) `The clientpatron relationship: emblem and reality in
Juvenal's rst book', in Patronage in Ancient Society, ed. A. Wallace-Hadrill
(LondonNew York) 20518.
Coarelli, F. (1983) Il foro romano, vol. I (Rome).
(1984) Ìside Capitolina, Clodio e i mercanti di schiavi', in Alessandria e il
mondo ellenistico-romano. Studi in onore di Achille Adriani, ed. N. Bonacasa,
A. Di Vita (Rome) 46175.
(1985) Il foro romano, vol. II (Rome).
(1997) Il Campo Marzio: dalle origini alla ne della repubblica (Rome).
Cornell, T. J. (1991) `Rome: the history of an anachronism', in City States in
Classical Antiquity and Medieval Italy, ed. A. Molho, K. Raaaub, J. Emlen
(Ann ArborStuttgart) 5369.
(1995) The Beginnings of Rome: Italy and Rome from the Bronze Age to the Punic
Wars (c.1000264 BC) (LondonNew York).
Crawford, M. H. (1974) Roman Republican Coinage (Cambridge).
(1996) ed. Roman Statutes (BICS Suppl. 64: London 1996).
D'Arms, J. H. (1970) Romans by the Bay of Naples (Cambridge, Mass.).
David, J.-M. (1973) et al., `Le `Commentariolum Petitionis' de Quintus
Ciceron: etat de la question et etude prosopographique', ANRW 1,3:
23977.
(1992) Le patronat judiciaire au dernier siecles de la republique romaine (Collec-
tion de l'ecole francais de Rome 277: ParisRome).
De Fine Licht, K. (1966) The Rotunda in Rome (Aarhus).
154 Bibliography
De Libero, L. (1992) Obstruktion: Politische Praktiken im Senat und in der
Volksversammlung der ausgehenden romischen Republik (Hermes Einzel-
schriften 59: Stuttgart).
De Ligt, L. (1993) Fairs and Markets in the Roman Empire (Amsterdam).
De Martino, F. (1989) Ìl modello della citta-stato', Storia di Roma 4 (Turin)
93358.
Demougin, S. (1987) `Quo descendat in campo petitor: Elections et electeurs a
la n de la republique et au debut de l'empire', in L'urbs: Espace urbain et
histoire I
er
avant J.-C. III
e
siecle apres J.-C. (Collection de l'ecole francais
de Rome 98: Rome) 30517.
Dench, E. (1995) From Barbarians to New Men (Oxford).
Deniaux, E. (1987) `De l'ambitio a l'ambitus: Les lieux de la propagande et de
la corruption electorale', in L'urbs: Espace urbain et histoire I
er
avant J.-C.
III
e
siecle apres J.-C. (Collection de l'ecole francais de Rome 98: Rome)
279304.
De Robertis, F. M. (1938) Il diritto associativo romano (Bari).
Develin, R. (1978) `Scipio Aemilianus and the Consular elections of 148 BC',
Latomus 37: 4848.
(1985) The Practice of Politics at Rome 366167 BC (Collection Latomus 188:
Brussels).
Di Gennaro G. (1993) Ì comizi centuriati di Cic. De re p. II. 22.3940:
attribuzione, struttura, giudizio politico', Athenaeum 71: 54565.
Doblhofer, G. (1990) Die Popularen der Jahre 11199 vor Christus (Vienna).
Drexhage, R. (1989) `Tabulae novae, frumentationes und die stadtro mischen
plebs', in Migratio et commutatio: Studien zur alten Geschichte und deren
Nachleben. Thomas Pekary zum 60. Geburtstag (St Katharinen) 11935.
Drummond, A. (1999) `Tribunes and tribunician programmes in 63 BC',
Athenaeum 87: 12167.
Earl, D. C. (1965) `Tiberius Gracchus' last assembly', Athenaeum 43: 95105.
Eder, W. (1991) `Who rules? Power and participation in Athens and Rome', in
City States in Classical Antiquity and Medieval Italy, ed. A. Molho,
K. Raaaub, J. Emlen (Ann ArborStuttgart) 16996.
Evans, R. J. (1997) `The Augustan ``purge'' of the senate and the census of 86
BC', Acta Classica 40: 7786.
Favory, F. (1976) `Classes dangereuses et crise de l'etat dans le discours
ciceronien', in Texte, Politique, Ideologie: Ciceron (Paris) 111233.
Fears, J. R. (1981) `Cult of virtues and Roman imperial ideology', ANRW
2.17.2 (1981) 827948.
Fezzi, L. (1997) `Lex Clodia de iure et tempore legum rogandarum', SCO 45:
297328.
Finley, M. I. (1983) Politics in the Ancient World (Cambridge).
Flambard, J.-M. (1977) `Clodius, les colleges, la plebe et les esclaves. Re-
cherches sur la politique populaire au milieu du I
er
siecle', MEFRA 89:
11556.
(1981) `Collegia Compitalicia: phenomene associatif, cadres territoriaux et
cadres civiques dans le monde romain a l'epoque republicaine', Ktema 6:
14366.
Flaig, E. (1995) Èntscheidung und Konsens: Zu den Feldern der politischen
Bibliography 155
Kommunikation zwischen Aristokratie und Plebs', in Demokratie in Rom?
Die Rolle des Volkes in der Politik der romischen Republik, ed. M. Jehne
(Historia Einzelschriften 96: Stuttgart) 77127.
Forbes, H. and L. Foxhall (1982) `Sitometreia: the role of grain as a staple food
in classical antiquity', Chiron 12: 4190.
Forsen, B. and G. Stanton, eds. (1996) The Pnyx in the History of Athens (Helsinki).
Fraschetti, A. (1990) Roma e il principe (BariRome).
Frier, B. W. (1980) Landlords and Tenants in Imperial Rome (Princeton).
Gabba, E. (1997) `Democrazia a Roma', Athenaeum 75: 26671.
Garnsey, P. D. A. (1981) Ìndependent freedmen and the economy of Roman
Italy under the Principate', Klio 63: 35971.
(1988) Famine and Food Supply in the Graeco-Roman World (Cambridge).
Gelzer, M. (1912) Die Nobilitat der romischen Republik (LeipzigBerlin).
Giovannini, A. (1995) `Catilina et le probleme des dettes', in Leaders and Masses
in the Roman World. Studies in Honor of Zvi Yavetz, ed I. Malkin and Z. W.
Rubinsohn (LeidenNew YorkCologne) 1532.
Grieve, L. J. (1985) `The reform of the comitia centuriata', Historia 34: 278309.
Gruen, E. S. (1974) The Last Generation of the Roman Republic (Berkeley).
(1991) `The exercise of power in the Roman republic', in City States in
Classical Antiquity and Medieval Italy, ed. A. Molho, K. Raaaub, J. Emlen
(Ann ArborStuttgart) 25267.
Hahn, I. (1975) `Der Klassenkampf der plebs urbana in den letzten Jahrzehnten
der ro mischen Republik', in Die Rolle der Volksmassen in der Geschichte der
vorkapitalistischen Gesellschaftsformationen, ed. J. Herrmann and I. Sellow
(Berlin) 43550.
Hall, U. (1964) `Voting procedure in Roman assemblies', Historia 13: 267306.
(1972) Àppian, Plutarch and the tribunician elections of 123 BC', Athenaeum
50: 335.
(1990) `Greeks and Romans and the secret ballot', in Òwls to Athens'. Essays
on Classical Subjects Presented to Sir Kenneth Dover, ed. E. M. Craik (Oxford)
1919.
(1998) ` ``Species libertatis'': voting procedure in the late Roman republic', in
Modus Operandi: Essays in Honour of Geoffrey Rickman, ed. M. Austin,
J. Harries, C. Smith (BICS Suppl. 71: London) 1530.
Hansen, M. H. (1987) The Athenian Assembly in the Age of Demosthenes (Oxford).
(1991) The Athenian Democracy in the Age of Demosthenes (Oxford).
(1993) `The battle exhortations in ancient historiography: fact or ction?',
Historia 42: 16180.
(1995) `The Athenian ecclesia and the Swiss Landesgemeinde', in Demo-
kratia, ed. K. H. Kinzl (Darmstadt) 32449.
(1996) `Reections on the number of citizens accommodated in the assembly
place of the Pnyx', in The Pnyx in the History of Athens, ed. B. Forsen and
G. Stanton (Helsinki) 2333.
Harris, W. V. (1989) Ancient Literacy (Cambridge, Mass.).
(1990) Òn dening the political culture of the Roman republic: some
comments on Rosenstein, Williamson, and North', CP 85: 28894.
Heikkila, K. (1993) `Lex non iure rogata: senate and annulment of laws in the
Late Republic', AIRF 13: 11742.
156 Bibliography
Hellegouarc'h, J. (1972) Le vocabulaire latin des relations et des partis politique sous
la republique (Paris) 2nd edn.
Hoffman, W. (1951) `Plebs', RE 21, 1: 73103.
Ho lkeskamp, K.-J. (1987) Die Entstehung der Nobilitat (Stuttgart).
(1995) Òratoris maxima scaena: Reden vor dem Volk in der politischen
Kultur der Republik', in Demokratie in Rom? Die Rolle des Volkes in der Politik
der romischen Republik, ed. M. Jehne (Historia Einzelschriften 96: Stuttgart)
1149.
Hopkins, K. (1978) Conquerors and Slaves: Sociological Studies in Roman History,
vol. I (Cambridge).
(1983) Death and Renewal: Sociological Studies in Roman History, vol. II (Cam-
bridge).
(1991) `From violence to blessing: symbols and rituals in ancient Rome', in
City States in Classical Antiquity and Medieval Italy, ed. A. Molho,
K. Raaaub, J. Emlen (Ann ArborStuttgart) 47998.
Jehne, M. (1993) `Geheime Abstimmung und Bindungswesen in der ro mischen
Republik', HZ: 593613.
(1995) `Die beeinuss von Entscheidungen durch ``Bestechung'': zur Funk-
tion des Ambitus in der ro mischen Republik', in Demokratie in Rom? Die
Rolle des Volkes in der Politik der romischen Republik, ed. M. Jehne (Historia
Einzelschriften 96: Stuttgart) 5176.
Johnson, T. and C. Dandeker (1989) `Patronage: relation and system', in Patronage
in Ancient Society, ed. A. Wallace-Hadrill (LondonNewYork) 21942.
Joshel, S. R. (1992) Work, Identity and Legal Status at Rome (Oklahoma).
Kornemann, E. (1900) `Collegium', RE 4,1: 380480.
Krause, C. (1981) `Per una riconstruzione graca del Comizio', PP 36: 712.
Kunkel, W. (1995) Staatsordnung und Staatspraxis der romischen Republik
(Munich).
Ku hnert, B. (1989) `Populus Romanus und sentina urbis: zur Terminologie der
plebs urbana der spaten Republik bei Cicero', Klio 71: 43241.
(1991) Die plebs urbana der spaten romischen Republik: Ihre okonomischen
Situation und soziale Struktur (Berlin).
La Rocca, E. (1995) `Sul Circo Flaminio', Arch. Laz. 12: 10319.
Laser, G. (1997) Populo et scaenae serviendum est: Die Bedeutung der stadtischen
Masse in der spaten Romischen Republik (Trier).
Last, H. ( 1932) `Tiberius Gracchus', CAH IX: 139.
Le Gall, J. (1971) `Rome, ville de faineants?', REL 49: 26677.
Liebenam, W. (1900) `Contio', RE 4,1: 114953.
(1903) `Divisor', RE 5: 12378.
Linderski, J. (1961) `Ciceros Rede pro Caelio und die Ambitus- und Vereinsge-
setzgebung der ausgehenden Republik', Hermes 89: 10619.
(1985) `Buying the vote: electoral corruption in the late republic', Anc. World
11: 8794.
Lintott, A. W. (1968) Violence in Republican Rome (Oxford).
(1987) `Democracy in the Middle Republic', ZSSR 104: 3452.
(1990) Èlectoral bribery in the Roman republic', JRS 80: 116.
(1994) `The crisis of the republic: sources and source-problems', CAH IX:
115.
Bibliography 157
Lo Cascio, E. (1994) `The size of the Roman population: Beloch and the
meaning of the Augustan census gures', JRS 84: 2340.
(1997) `Le procedure di recensus dalla tarda repubblica al tardo antico e il
calcolo della popolazione di Roma', in La Rome imperiale. Demographie et
logistique (Collection de l'ecole francais de Rome 230: Rome) 176.
Lotze, D. (1995) `Die Teilhabe des Bu rgers an Regierung und Rechtsprechung',
in Demokratia, ed. K. H. Kinzl (Darmstadt).
Luisi, N. D. (1995) `Sul problema delle tabelle di voto nelle votazioni legisla-
tive: contributo all'interpretazione di Cic. `àd Att.'' 1.14.5', Index 23:
41951.
McGoshin, P. (1994) Sallust: The Histories, vol. II (Oxford).
Mackie, N. (1992) `Popularis ideology and popular politics at Rome in the rst
centuries BC', RhM 135: 4973.
MacMullen, R. (1980) `How many Romans voted?', Athenaeum 58: 4547.
Marble, M. M. (1985) `Jury pay and assembly pay at Athens', in Crux: Essays
presented to G. E. M. de Ste Croix on his 75th birthday, ed. P. A. Cartledge,
F. D. Harvey (Exeter) 26597.
Marshall, B. A. (1997) `Libertas populi: the introduction of secret ballot in Rome
and its depiction on coinage', Antichthon 31: 5473.
Meier, C. (1956) `Praerogativa centuria', RE Suppl. 8: 56798.
(1965) `Populares', RE Suppl. 10: 549615.
(1980) Res publica amissa (Frankfurt am Main); 1st edn (Wiesbaden 1966).
Michels, A. K. (1967) The Calendar of the Roman Republic (Princeton).
Millar, F. (1984) `The political character of the classical Roman republic', JRS
74: 119.
(1986) `Politics, persuasion and the people before the Social War (15090
BC)', JRS 76: 111.
(1989) `Political power in mid-republican Rome: Curia or Comitium?', JRS
79: 13850.
(1995a) `Popular politics at Rome in the late republic', in Leaders and Masses
in the Roman World. Studies in Honor of Zvi Yavetz, ed. I. Malkin and Z. W.
(1995b) `The Roman libertus and civic freedom', Arethusa 28: 99105.
(1998) The Crowd in Rome in the Late Republic (Michigan).
Molho, A., K. Raaaub, J. Emlen, eds. (1991) City States in Classical Antiquity
and Medieval Italy (Ann ArborStuttgart).
Mo ller, C. (1993) `Die Mecennarii in der ro mischen Arbeitswelt', ZSSR 110:
296330.
Mommsen, Th. (1843) De collegiis et sodaliciis Romanorum (Kiel).
(18545) Romische Geschichte, 3 vols. (Leipzig).
(1887) Romisches Staatsrecht, 3 vols. (Leipzig).
Morel, J.-P. (1987) `La topographie de l'artisanat et du commerce dans la Rome
antique', in L'urbs. Espace urbain et histoire I
er
avant J.-C. III
e
siecle apres
J.-C. (Collection de l'ecole francais de Rome 98: Rome) 12755.
Morley, N. (1996) Metropolis and Hinterland: Rome and the Economy of Italy
(Cambridge).
Morstein-Marx, R. (1998) `Publicity, popularity and patronage in the Commen-
tariolum Petitionis', Class. Ant. 17: 25988.
158 Bibliography
Mouritsen, H. (1988) Elections, Magistrates and Municipal Elite: Studies in
Pompeian Epigraphy (ARID Suppl. 15: Rome).
(1998) Italian Unication: A Study in Ancient and Modern Historiography
(BICS Suppl. 70: London).
Muzzioli, M. P. (1995) Ì lavori per la Via Nazionale e il Diribitorium', Riv. Ist.
Naz. Arch. Stor. Arte 18: 13967.
Nadig, P. (1997) Ardet Ambitus: Untersuchungen zum Phanomen der Wahlbeste-
chungen in der romischen Republik (Frankfurt am Main).
Nardo, D. (1970) Il `Commentariolum Petitionis' (Padua).
Nicolet, C. (1970) `Le livre III des Res Rusticae de Varron et les allusions au
deroulement des comices tributes', REA 72: 12930.
(1973) `Polybe et les institutions romaines', Entretiens sur l'antiquite classique
20: 20958.
(1976) `Tesseres frumentaires et tesseres de vote', in L'Italie Preromaine et la
Rome Republicaine: Melanges offerts a Jacques Heurgon (Collection de l'ecole
francais de Rome 27) 70816.
(1977) Rome et la conquete du monde mediterraneen (Paris).
(1980) The World of the Citizen in Republican Rome (London) 1st edn (Paris
1976).
(1983) `Polybe et la ``constitution'' de Rome: aristocratie et democratie', in
Demokratia et aristokratia (Paris) 1535.
Nielsen, I. and B. Poulsen (1992) The Temple of Castor and Pollux, vol. I (Rome).
Nippel, W. (1980) Mischverfassungstheorie und Verfassungsrealitat in antike und
fruher Neuzeit (Stuttgart).
(1981) `Die Plebs urbana und die Rolle der Gewalt in der spaten ro mischen
Republik', in Von Elend der Handarbeit, ed. H. Mommsen and W. Schulze
(Stuttgart) 7092.
(1995) Public Order in Ancient Rome (Cambridge).
North, J. A. (1990) `Politics and aristocracy in the Roman republic', Past and
Present 126: 321.
Nowak, K. J. (1973) `Der Einsatz privater Garden in der spaten ro mischen
Republik' (Diss. Munich).
Packer, J. E. (1967) `Housing and population in imperial Ostia and Rome', JRS
57: 809.
(19689) `La Casa di Via Giulio Romano', Bull.Com. 81: 12748.
Palmer, R. E. A. (1971) `Tre lettere in cerca di storico', RFIC 99: 385409.
Palombi, D. (1997) Tra Palatino ed Esquilini: Velia, Carinae, Fagutal: storia
urbana di tre quartiere di Roma antica (Rome).
Pani, M. (1997) La politica in Roma antica: cultura e prassi (Rome).
Paterson, J. (1985) `Politics in the late republic', in Roman Political Life 90 BC
AD 69, ed. T. P. Wiseman (Exeter) 2143.
Patterson, J. R. (1994) `The collegia and the transformation of the towns in the
second century AD', in L'Italie d'Auguste a Diocletien (Collection de l'ecole
francais de Rome 198: Rome) 22738.
Perelli, L. (1982) Il movimento popolare nell'ultimo secolo della repubblica (Turin).
(1990) Il pensiero politico di Cicerone (Florence).
Picard, G.-Ch. (1965) `L'aedes libertatis de Clodius au Palatin', REL 43:
22037.
Bibliography 159
Pina Polo, F. (1995) `Procedures and functions of civil and military contiones in
Rome', Klio 77: 20316.
(1996) Contra arma verbis: Der Redner vor dem Volk in der spaten romischen
Republik (HABES 22: Heidelberg).
Prell, M. (1997) Armut im antiken Rom (Stuttgart).
Purcell, N. (1994) `The city of Rome and the plebs urbana in the late Republic',
CAH IX: 64488.
(1995) `Forum Romanum', Lex.Top.Urb.Rom. 2: 32536.
(1996) `Rome and its development under Augustus and his successors', CAH
X: 783811.
Ramage, E. S. (1987) The Nature and Purpose of Augustus' `Res Gestae', (Historia
Richardson, J. S. (1971) `The ``Commentariolum petitionis'' ', Historia 20:
43642.
Ritter, H.-W. (1998) `Zu Libertas und den Tabellargesetzen in der republikan-
ischen Mu nzpragung', in Imperium Romanum: Studien zu Geschichte und
Rezeption. Festschrift fur Karl Christ zum 75. Geburtstag ed. P. Kneissl,
V. Losemann (Stuttgart) 60814.
Rodriguez Almeida, E. (1980) Forma urbis marmorea: Aggiornamento generale
(Rome).
Rosenstein, N. (1993) `Competition and crisis in mid-Republican Rome',
Phoenix 47: 31338.
Rouland, N. (1981) Rome: Democratie impossible?
Royo, M. (1987) `Le quartier republicain du Palatin: nouvelles hypotheses de
localisation', REL 65: 89ff.
(1999) Domus Imperatoriae (BEFAR 303: Rome).
Ryan, F. X. (1994) `The reliability of Asconius In Cornelianam 69C', Riv. It.
Fil. Clas. 12: 1039.
Saller, R. (1982) Personal Patronage under the Early Empire (Cambridge).
Scheer, R. (1971) `Vindex libertatis', Gymnasium 78:1828.
Scheidel, W. (1994) `Libitina's bitter gains: seasonal mortality and endemic
disease in the ancient city of Rome', Anc. Soc. 25: 15175.
Schoeld, M. (1995) `Cicero's denition of Res Publica', in Cicero the Philosopher,
ed. J. G. F. Powell (Oxford) 6383.
Scobie, A. (1986) `Slum, sanitation, and mortality in the Roman world', Klio 68:
309433.
Scullard, H. H. (1973) Roman Politics, 220150 BC (Oxford) 2nd edn.
Seager, R. (1972) `Cicero and the word popularis', CQ 22: 32838.
Shackleton Bailey, D. R. (196570) Letters to Atticus, 7 vols. (Cambridge).
(1981) Cicero. Epistulae ad Quintum Fratrem et M. Brutum (Cambridge).
(1991) Cicero. Back From Exile: Six Speeches upon His Return (Chicago).
Shatzman, I. (1975) Senatorial Wealth and Roman Politics (Coll. Latomus 142:
Brussels).
Spielvogel, J. (1997) `P. Clodius Pulcher eine politische Ausnahmeerscheinung
der spaten Republik?', Hermes 125: 5874.
Stanton, G. (1996) `The shape and size of the Athenian assembly place in its
second phase', in The Pnyx in the History of Athens, ed. B. Forsen and
G. Stanton (Helsinki) 721.
160 Bibliography
Staveley, E. S. (1972) Greek and Roman Voting and Elections (London).
Stockton, D. (1971) Cicero: A Political Biography (Oxford).
Storey, G. R. (1997) `The population of ancient Rome', Antiquity 71: 96678.
Stylow, U. A. (1972) `Libertas und Liberalitas: Untersuchungen zur innenpoli-
tischen Propaganda der Ro mer' (Diss. Munich).
Syme, R. (1939) The Roman Revolution (Oxford).
Taylor, L. R. (1949) Party Politics in the Age of Caesar (Berkeley).
(1960) The Voting Districts of the Roman Republic (Rome).
(1962) `Forerunners of the Gracchi', JRS 52: 1927.
(1963) `Was Tiberius Gracchus' last assembly electoral or legislative?',
Athenaeum 41: 3969.
(1966a) Roman Voting Assemblies (Ann Arbor).
(1966b) Àppian and Plutarch on Tiberius Gracchus' last assembly',
Athenaeum 44: 23850.
Thommen, L. (1989) Das Volkstribunat der spaten romischen Republik (Historia
(1995) `Les lieux de la plebe et de ses tribuns dans la Rome republicaine', Klio
77: 35870.
Treggiari, S. (1969) Roman Freedmen during the Late Republic (Oxford).
(1980) Ùrban labour in Rome: mercennarii and tabernarii', in Non-slave
Labour in the Graeco-Roman World, ed. P. Garnsey (PCPS Suppl. 6: Cam-
bridge) 4864.
Ulrichs, R. B. (1994) The Roman Orator and the Sacred Stage (Collection
Latomus 222: Brussels).
Vaathera, J. (1993) Òn the religious nature of the place of assembly', AIRF 13:
97116.
Vanderbroeck, P. J. J. (1987) Popular Leadership and Collective Behavior in the
Late Roman Republic (ca. 8050 BC) (Amsterdam).
Venturini, C. (1984) `L'orazione pro Plancio e la lex Licinia', in Studi in onore di
Cesare Sanlippo, 8 vols. (198291), (Milan) 5: 789804.
Veyne, P. (1990) Bread and Circuses (London), 1st edn (Paris 1976).
Virlouvet, C. (1985) Famines et emeutes a Rome des origines de la republique a la
mort de Neron (Collection de l'ecole francais de Rome 87: ParisRome).
(1995) Tessera Frumentaria: Les procedures de distribution du ble public a la n de
la republique et au debut de l'empire (BEFAR 286: Rome).
(1996) Ùne allusion varronienne aux fraudes de Clodius? A
propos de res
rusticae III.5.18', MEFRA 108: 873901.
Walbank, F. W. (1972) Polybius (Berkeley).
(1995) `Polybius' perception of the one and the many', in Leaders and Masses
in the Roman World. Studies in Honor of Zvi Yavetz, ed. I. Malkin and Z. W.
Wallace-Hadrill, A. (1989) `Patronage in Roman society: from republic to
empire', in Patronage in the Ancient World, ed. A. Wallace-Hadrill (London
New York) 6387.
Wallinga, T. (1994) Àmbitus in the Roman world', Rev. Int. Dr. Ant. 41:
41142.
Waltzing, J.-P. (18951900) Etude historique sur les corporations professionelles chez
les Romains, 4 vols. (Louvain).
Bibliography 161
Whittaker, C. R. (1993) `The poor', in The Romans, ed. A. Giardina (Chicago
London) 28299.
Will, W. (1991) Der romische Mob (Darmstadt).
Wirszubski, C. (1950) Libertas as a Political Idea at Rome during the Late Republic
and Early Principate (Cambridge).
Wiseman, T. P. (1969) `The census in the rst century BC', JRS 59: 5975.
(1971) New Men in the Roman Senate 139 BCAD 14 (Oxford).
(1974) `The Circus Flaminius', PBSR 42: 226.
(1985) `Competition and co-operation', in Roman Political Life 90 BCAD 69,
ed. T. P. Wiseman (Exeter) 319.
(1994) `The senate and the populares', CAH IX: 32767.
(1999) `Democracy alla romana', JRA 12: 53740.
Wood, N. (1988) Cicero's Social and Political Thought (Berkeley).
Yakobson, A. (1992) `Petitio and largitio: popular participation in the centuriate
assembly of the late republic', JRS 82: 3252.
(1995) `Secret ballot and its effect in the late Roman republic', Hermes 123:
42642.
(1999) Elections and Electioneering in Rome: A Study in the Political System of the
Late Republic (Historia Einzelschriften 128: Stuttgart).
Yavetz, Z. (1958) `The living conditions of the urban plebs in republican Rome',
Latomus 17: 50017.
(1969) Plebs and Princeps (Oxford).
Ziolkowski, A. (1996) Òf streets and crossroads: the location of the Carinae',
MAAR 41: 12151.
Index
M. Aemilius Lepidus (cos. 78) 51, 66
L. Aemilius Paullus (cos. 182) 64, 71
M. Aemilius Scaurus (pr. 56) 108
T. Annius Milo 47, 49, 62, 11213, 115
C. Antonius Hybrida (cos. 63) 104
M. Antonius (triumvir) 122 n.94
L. Appuleius Saturninus (tr. 100) 6971,
82, 87
C. Asinius Pollio (cos. 40) 72
Aristotle 8, 130
Athens 356, 46, 934, 129
M. Audius Lurco (tr. 61) 113
Augustus 911, 26, 36 n.72
Balsdon, J. P. V. D. 4 n.11
Bernstein, A. H. 67
boni 41, 445, 52, 55, 62, 71, 74 n.29, 79,
134, 140, 1478
bribery 98110, 11213, 114, 118,
1245, 144
Brunt, P. A. 98
Burton, G. 13
Q. Caecilius Metellus Nepos (tr. 62) 53
Q. Caecilius Metellus Numidicus (cos.
109) 87
L. Caecilius Rufus (tr. 63) 55, 66
M. Calpurnius Bibulus (cos. 59) 45, 49, 54
n.44, 58
C. Calpurnius Piso (cos. 67) 48
Capitol 18, 256, 40, 64
Carafa, P. 19
Cassius Sabaco 29
Circus Flaminius 18, 26, 40
Cisalpine Gaul 1201
Ap. Claudius Pulcher (cos. 54) 17 n.54, 50
clientela 3, 64, 6778, 96100, 102, 125,
1378
P. Clodius Pulcher 9, 21, 25 n.24, 33, 42,
4752, 54, 5661, 69, 80, 83, 856,
88, 101, 122, 142, 14951
Sex. Cloelius 83, 85
Coarelli, F. 18, 25, 35
C. Coelius Caldus (cos. 94) 69
coitio 103
collegia 58, 834, 142
comitia centuriata 10, 269, 34, 64 n.4, 66,
83, 88 n.66, 94, 11415, 120
comitia curiata 34
comitia tributa/concilium plebis 26, 303, 63,
94
Comitium 1820, 35
Compitalicia 83, 142
contio 1315, 17, 20, 23, 246, 32, 34,
3863, 67 n.13, 79, 131
C. Cornelius (tr. 67) 48, 58, 66, 69, 85
Cn. Cornelius Lentulus Marcellinus (cos.
56) 51, 54, 61
P. Cornelius Lentulus Spinther (cos. 57)
567
P. Cornelius Lentulus Sura, Catilinarian 58
P. Cornelius Scipio Aemilianus (cos. 147)
7, 25, 47 n.27, 53, 59, 77, 79, 83, 98,
101, 124
P. Cornelius Scipio Nasica (cos. 138) 47,
53
L. Cornelius Sulla 51, 6970, 126, 146
Cozza, L. 28
curatores tribuum 32, 113
Curia 1820, 24
democracy 3, 68, 1416, 389, 46, 63,
74, 90 n.1, 100, 1448
divisores 110, 11314, 143 n.37, 1501
Cn. Domitius Ahenobarbus (cos. 96) 69
Cn. Domitius Calvinus (cos. 53) 97, 99
donations 97, 109, 11112, 143
C. Fannius (cos. 122) 67 n.13
M. Favonius (pr. 49) 23 n.19, 110
C. Flaminius (cos. 223) 73, 79
L. Flavius (tr. 60) 52 n.42
freedmen 34, 401, 59, 63, 70, 72, 77, 80,
85, 94
162
Index 163
Forma urbis 279
Forum Romanum 18, 201, 24, 435, 50,
58, 70, 107, 124, 142 n.35
Q. Fuus Calenus (cos. 47) 47 n.26, 54
M. Fulvius Flaccus (cos. 125) 66
A. Gabinius (cos. 58) 49, 69
games 97, 111, 125 n.102
Garnsey, P. D. A. 40
grain dole 4, 32, 36, 856, 133, 146
Gruen, E. S. 86
Hall, U. 103
Hansen, M. H. 19 n.4, 25
homines novi 35
Hopkins, K. 12
immigration 72, 74, 801
Italians 33, 74 n.29, 95, 11823
iustitium 59
C. Julius Caesar 1, 9, 26, 30, 44, 49, 54
n.44, 58, 69, 97, 112, 121 n.94, 136
M. Junius Brutus (tr. 83) 70
M. Junius Pennus (tr. 126) 80
M. Juventius Laterensis (pr. 51) 31, 91, 98,
119
C. Laelius (cos. 140) 66
Leges Aelia et Fua 34
Leges Clodiae 69, 142 n.36, 146
Lex Baebia 124
Lex Caecilia Didia 87
Lex Calpurnia de ambitu 110
Lex Calpurnia de repetundis 68, 73, 76
Lex Cassia tabellaria 68
Lex Cincia 138 n.23
Lex Cornelia Fulvia 124
Lex Fabia 107
Lex Gabinia de imperio Cn. Pompei 13 n.44,
26 n.28, 47, 52
Lex Gabinia tabellaria 68, 73, 756, 79
Lex Hortensia 73
Lex Licinia de sodalitatibus 88, 114, 14951
Lex Licinia Mucia 80
Lex Manilia de provinciis Cn. Pompei 117
Lex Octavia frumentaria 87, 146
Lex Oppia 52, 78
Lex Papiria tabellaria 82
Lex Poetilia 35
Lex Porcia frumentaria 147
Lex Rubria 53
Lex Servilia 87
Lex Terentia Cassia 146
libertas 914, 46, 131
C. Licinius Crassus (tr. 145) 20, 24, 49, 64
C. Licinius Crassus (cos. 70) 51, 54, 60
n.65, 62
L. Licinius Murena (cos. 62) 71, 1223
P. Licinius Nerva (pr. 104) 212
M. Livius Drusus (tr. 122) 123, 65, 85
M. Livius Drusus (tr. 91) 48, 66, 70, 71
n.23, 856
L. Lucceius (pr. 67) 110 n.57, 143
Q. Lutatius Catulus (cos. 78) 47
MacMullen, R. 21, 23, 28, 32
C. Manilius (tr. 66) 58, 66, 6970, 142
n.36
L. Marcius Philippus (cos. 91) 48, 53, 65,
87
C. Marius (cos. 107) 22 n.14, 29, 367,
69, 71, 76, 79, 82, 96 n.12, 98, 101
M. Marius Gratidianus (pr. 85) 83
Meier, C. 39, 61
C. Memmius (pr. 58) 99
Millar, F. 35, 1314, 16, 74, 90 n.1
Minucius Rufus (tr. 121) 53, 878
Mommsen, T. 39, 61, 64
T. Munatius Plancus (tr. 52) 49
Nicolet, C. 30, 1034
nomenclatores 107, 125
North, J. A. 89
M. Octavius (tr. 133) 43, 62, 81
oratory, public 13, 556, 74, 78
Cn. Papirius Carbo (cos. 120) 53, 64,
789
Patterson, J. R. 84
Cn. Plancius (tr. 56) 31, 103, 114,
11920, 122, 14950
Plautus 41, 434
plebs contionalis 3943, 57, 78, 87
A. Plotius (pr. 51) 103, 119
Polybius 57, 124
Pompeii 104, 136
Cn. Pompeius Magnus 9 n.30, 30, 41, 45,
514, 69, 71 n.20, 87, 92 n.4, 97,
101, 117, 136
pontes 19, 213
P. Popilius Laenas (cos. 132) 87
populares 2, 10, 123, 64, 6771, 73, 81,
93, 1312
population of Rome 32, 133
C. Porcius Cato (tr. 56) 53, 72
M. Porcius Cato Maior (cos. 195)
M. Porcius Cato Uticensis (pr. 54) 22
n.15, 23 n.19, 52, 61, 98, 146
the poor 59, 767, 79, 82, 1347,
13941
164 Index
praerogativa centuria 31, 99, 1012, 106,
115
Rostra 18, 24
rural voters 345, 72, 78, 804, 91, 95
Saepta 2630, 91
Saller, R. 137
salutatio 75, 77n.41, 107, 109
C. Scribonius Curio (cos. 76) 58
C. Scribonius Curio (tr. 50) 44, 72, 92
n.4
sectatores 767, 107, 109, 116, 125
C. Sempronius Gracchus 9, 245, 44, 49,
65, 67 n.13, 689, 71, 76, 802, 85,
878, 91, 101, 136, 146
Ti. Sempronius Gracchus 9, 22 n.15, 23,
43, 53, 57, 68, 71, 812, 84
Senate meetings 33, 48 n.33
senatus consultum ultimum 69, 148
sequestri 11314, 1501
L. Sergius Catilina (pr. 68) 9, 589, 104,
142 n.36, 147
Sergius 59, 140
Q. Servilius Caepio (pr. 91) 71
C. Servilius Glaucia (tr. 101) 69
P. Servilius Rullus (tr. 63) 14, 534, 656,
69, 92 n.4
Sicinius (tr. 76) 51 n.40
sodalitates 110 n.58, 114, 144, 14951
P. Sulpicius (tr. 88) 6970
P. Sulpicius Galba (pr. 66) 106
Ser. Sulpicius Galba (cos. 144) 64, 78
Ser. Sulpicius Galba (pr. 54) 73, 78
Syme, R. 2
tabernarii 42, 59, 108, 140
Taylor, L. R. 21, 28, 32
Temple of Castor 215
Q. Terentius Culleo (pr. 187) 73
Sex. Titius (tr. 99) 6970
C. Trebellius (tr. 67) 69
Tribunal Aurelium 58, 142 n.35
tribules 96, 109
tribus 1826, 801, 1034, 10915, 119,
122
urban 34, 40, 63, 801
M. Tullius Cicero (cos. 63) passim
consular election 93, 95, 105, 106 n.45,
1079, 1203
exile and restoration 9, 33, 567, 88,
109 n.53, 121
optimate ideology 911, 129 n.1,
1401
C. Valerius Tappo (tr. 188) 73
Vanderbroeck, P. J. J. 4, 401, 72
P. Vatinius (tr. 59) 69, 120 n.88
C. Verres (pr. 74) 114
vici/vicini 71, 834, 96, 1412
voting procedures 19, 2131
duration 223, 26 n.29, 301, 1034,
105 n.42,
Wallace-Hadrill, A. 67, 74
Wiseman, T. P. 25
written ballot 26, 37 n.75, 68, 756, 79,
111
Yakobson, A. 4, 95, 112
Yavetz, Z. 147

Henrik Mouritsen-Plebs and Politics in The Late Roman Republic-Cambridge University Press (2001)

Uploaded by

Copyright:

Available Formats

Henrik Mouritsen-Plebs and Politics in The Late Roman Republic-Cambridge University Press (2001)

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Henrik Mouritsen-Plebs and Politics in The Late Roman Republic-Cambridge University Press (2001)

Uploaded by

Copyright:

Available Formats

This page intentionally left blank

You might also like