At some point between the first two Punic Wars, the comitia centuriata was reformed.Footnote 1 This venerable voting assembly was distinguished from other electoral institutions at Rome by an overtly timocratic structure that gave the richest Romans a disproportionately high share of the vote and had them clinch a majority before poorer Romans even had a chance to participate. According to the old ‘Servian’ system—supposedly designed by the penultimate king, Servius Tullius—the total votes of the cavalrymen and of the richest infantry would together provide a majority of votes to seal a result: ninety-eight out of one hundred and ninety-three. After the reform, however, the balance changed. Now these two wealthy groups could only command eighty-eight out of one hundred and ninety-three centuries, or nine short of the critical majority of ninety-seven.
All electoral politics is ultimately mathematical, and the rejigged numbers of the comitia centuriata have long tantalized historians. To reapportion votes in a timocratic assembly smacks of democratic or oligarchic politics, and it would not take an Aristotle or a Marx to understand that robbing the landed elites of a majority vote has the feel of democratic or proletarian reform. Dionysius of Halicarnassus claimed that the reform pushed the assembly in a more democratic direction (εἰς τὸ δημοτικώτερον), while De Martino saw the reform as part of ‘le tendenze popolari’, which sought to make Rome ‘più popolare’.Footnote 2 But others have rejected this.Footnote 3 Democratization does not sit easily with Rome's hierarchical political culture, and in any case the numbers were only tweaked; there was no attempt to distribute the proletarii, no slashing of the elite vote and no allowance of poorer voters to vote alongside or even before the wealthy. If this was an attack on the voting power of the rich, it was an exceedingly light blow.
As Meier noted in 1956, the sheer volume of theories is a mark of the uncertainty that shrouds this reform; we are unlikely ever to know for certain what the reformers had in mind.Footnote 4 Yet aspects of this reform have not been properly explored. They shed revealing light on the issue and provide a unifying rationale for a reform that had numerous parts. The contention of this article is that the reforms were not designed to recalibrate the voting power of the rich relative to the rest of the community; to the extent that there were such arithmetical effects, they were incidental. Instead, the reform was intended to reapportion voting power within the elite after the trials of the First Punic War, and the basis for that reapportionment was not so much timocracy as it was geography.
WHAT WERE THE REFORMS?
By the time of the reform, the comitia centuriata was already a venerable institution (Livy 1.43.12, Dion. Hal. Ant. Rom. 4.16–18). Divided into one hundred and ninety-three individual centuries, each century was conceptualized as a part of the army, with eighteen centuries reserved for the equites, one each for specialists such as musicians and fabri tignarii, one for the landless who did not qualify for military service, and then an overwhelming majority for the infantry divided into five classes by wealth.Footnote 5 In its pre-reform guise, the eighteen cavalry centuries voted first, and then the eighty centuries of the first class; the arithmetic logic is clear from the fact that unanimity between these two groups would yield a majority vote of ninety-eight centuries out of one hundred and ninety-three (Cic. Rep. 2.39). If these richest Romans all agreed, then the election was over, and the poorer Romans would never even be called to vote. In the third century, however, the assembly was reorganized. The precise date of the reform is unknown, but it must have been between 241 and 218 b.c.e. By the time of the latter date, when Livy's Book 21 begins, the changes were in place. And the reform must have post-dated the creation of the last two tribes in 241, since the reform depended on the existence of thirty-five tribes in total.
No extant source explicitly discusses the reform—Livy may have done so in his lost second decade—but passing mentions leave a detailed enough impression. The evidence is a combination of antiquarian discussion of the earlier ‘Servian’ assembly and a set of descriptions of how the post-reform assembly functioned; contrasting the operations of the earlier and the later comitia establishes what the reform actually changed. Livy, discussing Servius Tullius’ institutions at 1.43.12, displays his antiquarian learning:
equites enim uocabantur primi; octoginta inde primae classis centuriae; ibi si uariaret, quod raro incidebat, institutum ut secundae classis uocarentur, nec fere unquam infra ita descenderunt, ut ad infimos peruenirent. nec mirari oportet hunc ordinem, qui nunc est post expletas quinque et triginta tribus duplicato earum numero centuriis iuniorum seniorumque, ad institutam ab Ser. Tullio summam non conuenire.
For the equites used to be called first; then the eighty centuries of the first class: if there were any disagreement there, which used to occur rarely, it was instituted that the centuries of the second class would be called; and almost never did they descend so far that they reached the poor. Nor should it surprise anyone that the system which exists now, since the tribes reached thirty-five and their number was doubled by means of centuries of iuniores and seniores, does not match the total established by Servius Tullius.Footnote 6
Livy's readers knew how the post-reform comitia worked. His task, then, was to establish that the old assembly was different in that it used to call cavalry before infantry, that there used to be eighty centuries in the first class of infantry, that the reason Late Republicans only had seventy centuries in the first class was that (an undated) reform granted two centuries (one for the iuniores and one for the seniores) to each of the thirty-five tribes. The use of the imperfect tense—‘used to be called first’—emphasizes that the old way was no longer in use. And so, although no extant passage of Livy addresses the moment of reform, it is clear that it shifted the voting order of the equites and redistributed centuries in the infantry by assigning them on the basis of an age division in the tribes. Each of those features can be addressed in turn.
The equites used to vote before the first class of the infantry, but passing references reveal the nature of that change. Describing an election in 44 b.c.e., Cicero (Phil. 2.82) seems to indicate that the six most esteemed centuries of equestrians (the sex suffragia) were separated from the rest of the cavalry and began to vote between the first and the second classes, though frustratingly the text refers only to suffragia rather than to sex suffragia:
ecce Dolabellae comitiorum dies. sortitio praerogatiuae: quiescit. renuntiatur: tacet. prima classis uocatur [renuntiatur], deinde, ita ut adsolet, suffragia, tum secunda classis, quae omnia sunt citius facta quam dixi.
Look at the day of Dolabella's election. The centuria praerogatiua is drawn by lot. He does nothing. It is announced. He says nothing. The first class is called, then, as is usual, the suffragia, then the second class, all of which happened more quickly than I have reported it.
The abbreviated paratactic style of the passage makes it easy to believe that Cicero omitted the word sex from suffragia for his audience of cognoscenti.Footnote 7 The impression left by this passage is that the sex suffragia voted after the first class, while the other twelve equestrian centuries voted alongside the first class. This impression is reinforced at Rep. 2.39 (discussed below), in which Cicero, while counting up the centuries, distinguishes the twelve equestrian centuries from the sex suffragia. The clearest confirmation, however, comes from a passage of Livy, which recounts the unsuccessful trial of a censor, Ap. Claudius Pulcher, accused in 169 of abusing a tribune's powers in a dispute about public contracts. In this case, Livy implies that only twelve of the eighteen centuries were eligible to vote alongside the first class. Moreover, it was only after the votes of the first class and at least eight of the twelve equestrian centuries that the political leaders strained to win over the plebs that had not yet voted (43.16.14):
cum ex duodecim centuriis equitum octo censorem condemnassent multaeque aliae primae classis, extemplo principes ciuitatis in conspectu populi anulis aureis positis uestem mutarunt, ut supplices plebem circumirent.
When eight of the twelve centuries of equites and many others of the first class had voted to convict the censor, the most eminent men of the city immediately set aside their gold rings in full view of the people and put on mourning, so that they could move among the plebs as beseechers.
The clear implication of Livy's description is that only twelve equestrian centuries were expected to vote together with the first class. Six others—the sex suffragia—were to vote later, only after the first class had issued its judgement. Hence twelve equestrian centuries voted alongside the first class, while the sex suffragia voted between the first and the second classes.
The next issue raised at Livy 1.43.12 above is the allocation of two centuries to each of the thirty-five tribes. The passage never specifies that the doubling of the thirty-five tribes was limited to the first class, and for some time this spurred lively debate.Footnote 8 The problem, however, is that, if all five classes of the comitia centuriata had seventy centuries, then classes 2–5 would swell from the ninety centuries of the conventional version to two hundred and eighty centuries. This faces the insurmountable evidence of Cicero (Rep. 2.39) that there were only one hundred and ninety-three centuries in total:
nunc rationem uidetis esse talem, ut equitum centuriae cum sex suffragiis et prima classis addita centuria, quae ad summum usum urbis fabris tignariis est data, octoginta nouem centurias habeat; quibus ex centum quattuor centuriis (tot enim reliquae sunt) octo solae si accesserunt, confecta est uis populi uniuersa.
You already see such a rationale, so that the centuries of equites, with the sex suffragia and the first class, and with the addition of that century which has been reserved for the army engineers because they are of such great use to the city, would have eighty-nine centuries in total. If just eight out of the other one hundred and four centuries (for that is how many remain) agreed, the whole power of the people was fulfilled.
Despite valiant attempts, it has been impossible to overcome Cicero's statement here that there were only one hundred and four centuries to share between the remaining classes. Moreover, Staveley and Grieve each noted that when a tribe's century of iuniores or seniores is identified the class is never mentioned; we read, for example, of the Votinia iuniorum but not of the Votinia iuniorum of this or that class. This must imply that only one class from each tribe was split into iuniores and seniores.Footnote 9 Otherwise, our sources would need to specify not only the age bracket and the tribe, but which of the five possible classes the century belonged to as well. The fact that they do not mention the class implies that there was only one class that had centuries of iuniores and seniores: the first class.
As Staveley showed long ago, therefore, the sources reveal a fairly clear portrait of how the assembly worked after the reform.Footnote 10 There were indeed one hundred and ninety-three centuries, and only the centuries of the first class were distributed two per tribe. Of these, one century was drawn by lot to vote first as the centuria praerogatiua, and every such century that is named in our sources is a century of iuniores.Footnote 11 As for voting order, Cicero (Phil. 2.82) shows that it proceeded from the praerogatiua through the first class and twelve equites centuries, then on to the sex suffragia and the second class (with the fabri tignarii included somewhere in this mix as well).Footnote 12 The remaining classes followed.
The reform, therefore, comprised four steps, and any explanation for the reform must encompass them all:
1. The equestrian centuries lost the right to begin the voting. Twelve of the eighteen equestrian centuries would vote along with the first class of infantry, but the six most prestigious equestrian centuries (the sex suffragia), which included the senators, now voted between the first class of the infantry and the second class of the infantry.Footnote 13
2. The centuries of the first class were reorganised so that each of the thirty-five tribes received one century for the iuniores (under 45) and one for the seniores (over 45).
3. One century from the first class would be drawn by lot to initiate the vote, and this century almost certainly had to be of iuniores. This century was (or possibly came to be) termed the centuria praerogatiua.
4. The reduction of the first class from eighty to seventy centuries meant that ten centuries had to be absorbed by another class, but how that reassignment was managed is unknown. Those lower classes were not linked to the tribes.
The reforms themselves, then, appear secure enough. Left unclear is the point of this reform, and the challenge is to find an explanation that addresses all of the changes. Explanations fall into three main camps.
The first is the democratic explanation. This argues that the purpose was (1) to reduce the voting power of the most distinguished by forcing them to vote after the first class had voted; (2) to reduce the voting power of the first class by restricting the number of their centuries; and (3) to expand the voting power of the lower classes by requiring at least some votes from the second class before a majority of centuries could be reached.Footnote 14 This is not all that convincing. The relegation of the sex suffragia was no doubt a blow to the most privileged voters, but they were still voting in the first half, before others could secure a majority of votes. C. Gracchus apparently imagined a much more democratic reform that established a voting order by lot across all one hundred and ninety-three centuries, and the third century reform was certainly not that.Footnote 15 Nor was there any reduction in the number of equestrian centuries; it is even possible that the property requirement of the equestrian centuries became more exclusive in this period to strengthen the vote of the wealthiest Romans.Footnote 16 But it is the reform to the first class that undermines any claim that this was reform in favour of the non-elite. True, the change in number did cross the critical threshold of a majority, but this is less important than it might at first seem. First, the reduction was only by ten centuries; a true democrat could theoretically have reduced it by twenty or thirty or abolished the timocratic classes altogether.Footnote 17 Moreover, the importance of the bare majority only matters if we accept that the equestrians and the first class voted identically. There is no reason to think that this was common; Livy's use of the imperfect tense (quod raro incidebat) above suggests that a split among the elite used to be rare, but had become normal, while Dionysius of Halicarnassus suggests that voting routinely reached the second and the third classes.Footnote 18 It is likely that Livy's portrait of landslide voting is little more than a fantasy of harmonious ancestors. If voting usually reached into the second or the third classes, then whether the wealthy centuries came to eighty-eight or ninety-eight was not decisive, since early voting rarely clinched a unanimous majority anyway. Nor is there clear evidence that Roman voters of the first class voted differently from voters in other classes. Voters of the first and the third classes may have assessed candidates and questions along the same lines and without the sorts of political class filters ascribed to modern voters.Footnote 19 The final problem with the idea that reform was designed to reduce the vote for the first class is that it fails to explain why the centuries of the first class were linked to the tribes. If reformers had wanted seventy centuries of the first class, they could have just moved ten centuries to another class. Or they could have picked sixty centuries, or fifty centuries. The key to the reform was not the number but the redistribution of centuries among the tribes. The goal, I will argue, was not to reduce the centuries of the first class by ten but to distribute centuries of the first class evenly across the tribes.
The democratic argument is problematic because the reform barely affected voting power; it fails to explain the central change of incorporating the tribes into the comitia centuriata. The second camp of explanations seizes upon this tribal reform to claim that the reform was designed precisely to bring the assembly more in line with Rome's reliance on tribes for many administrative functions. Develin, for example, argues that the reform was designed to link tribes with centuries so that conscription was more easily spread around.Footnote 20 This, however, makes no sense once we accept that one class of infantry was given such treatment and not the others or the cavalry; they were all subject to conscription, too. Nicholls similarly argued that the tribal rolls held voters’ details, and so it was easier to use one list to register people into both tribes and centuries; but he acknowledged that this only makes sense if we accept that all five classes were given two centuries per tribe, which they were not.Footnote 21 Administrative reform seems a weak explanation if it applied to only one class of citizens.
The third camp focusses on the creation of the centuria praerogatiua as a goal. This institution drew by lot one century from the first class to vote first, and its choice undoubtedly exercised great influence over subsequent centuries.Footnote 22 If one accepts that the vote of this praerogatiua was more or less decisive, then the election of consuls owed as much to the result of luck (that is, which century was drawn) as it did to a truly political decision. This offered the advantage that losing candidates could console themselves that they were rejected by an uncontrollable factor such as luck or fate, rather than the esteem in which they were held.Footnote 23 It also prevented any potentially heated conflict to win the votes of later centuries if the result was tight; once the praerogatiua had more or less announced the winner, the result was a fait accompli. Moreover, a tendency to follow the praerogatiua would lead to landslide elections that disguised any lack of real consensus among voters.Footnote 24 If it is true that the vote of the centuria praerogatiua really did hold this sort of sway over subsequent voters, all this would indeed be helpful in maintaining elite consensus.
Meier understood, however, that this could not be the (chief) purpose of the reform agenda unless it could explain all of the features of the reform. He then exerted himself to explain how the relegation of the sex suffragia after the first class—perhaps the only unambiguously anti-oligarchic part of the whole programmeFootnote 25—could have been part of this goal. However, his conclusion that this position was influential is difficult to accept, given the logic of the rest of the comitia; in every other part of the assembly, voting early is a privilege, so that the first class voted before the second, the second before the third, and so on.Footnote 26 And there are bigger problems, which leave other features of the reform inexplicable. If the goal was to endow the first century with the authority of sortitio, for example, why exclude the sex suffragia? Furthermore, why link the first class to tribes, since one can easily imagine the old Servian system with one of the first ninety-eight centuries drawn by lot? Mouritsen has identified this last problem of the tribal link. He argues that, in a Rome that was increasingly tying citizenship to tribes, the reform strengthened the authority of the assembly, but he understands that this is only a partial explanation.Footnote 27 That thesis does not address the voting order, as Mouritsen acknowledges, but it also does not address why the tribal association would be limited to the first class. Why not the equites and the lower classes, too? Would a complete tribal overhaul not strengthen the authority of the assembly even more? And then there is the problem of the iuniores. Although there is no decisive proof, every known centuria praerogatiua is of iuniores, and it is probably true that only the iuniores were eligible.Footnote 28 Taylor strengthened this view by accepting that the Latin expression ‘to throw the sixty-year-olds from the bridge’ (sexagenarios de ponte deicere) was a remnant of this promotion of iuniores voters, since the voting platform was known as a ‘bridge’.Footnote 29 If the goal was simply to establish an authoritative first century by lot, it is not clear why the centuries had to be based on tribes, why the praerogatiua had to be from the iuniores, or why it would have occurred to anyone that age was a meaningful category in the first place. Even if we accept, therefore, that the creation of the centuria praerogatiua was a goal of the reform, there is so much left unexplained:
— Why were the most illustrious Romans (those in the sex suffragia) banned from being drawn as the praerogatiua?
— Why did the whole first class have to vote before the sex suffragia, instead of allowing the luminaries to vote among the rest of the cavalry with the first class or perhaps even sandwiching the sex suffragia between the praerogatiua and the rest of the first class?
— Given that six of the equestrian centuries were relegated in this way, why were the other twelve apparently not?
— Why did the first class need to be associated with tribes?
— Why was the first class the only class to be distributed among tribes?Footnote 30
— Why was there a division between iuniores and seniores?
— Why was the praerogatiua apparently always drawn from the iuniores?
It is difficult to examine the full suite of reforms and conclude that all this was merely to create a bandwagon effect that would avoid splitting the vote. There are so many details that target such specific demographics. The demarcation of different groups—young vs old, first class vs other classes, sex suffragia vs other equestrians—implies that political rebalancing was a leading consideration. The winners and the losers are too clearly delineated for this not to be about the reapportionment of political power. And so we are left with the usual question: cui bono? The biggest winners in all this were, I argue, the iuniores who lived furthest from Rome. To explain why—and why it matters—requires a clearer understanding of what ‘tribalizing’ the centuries really meant.
THINKING WITH TRIBES
To align centuries with tribes was almost literally to map the comitia centuriata onto the various regions of central Italy. Although the tribes were technically agglomerations of people rather than territories, in practice they had a strong topographical sense—Hannibal, for example, marched in Pupiniam when he entered the territory of the Pupinian tribe.Footnote 31 With the reforms of the comitia centuriata, geography became central to the way in which the votes of the first class were distributed.Footnote 32 Because each tribe represented a specific area, the voters of the first class of each area were now guaranteed equality, with one collective comitial vote for the iuniores and one for the seniores. The point of the reform of the first class was not to create a specific number of first-class centuries but to embed centuries in specific communities in specific places. The number seventy was not a goal of reform; it was simply the unavoidable corollary of creating territory-based centuries in the first class.
Why, however, would the Romans shift from a purely timocratic organization for the comitia to a partially territorial one? And why would it only affect the first class? To answer this, it is worth examining how the voters from different regions would fare in the pre- and in the post-reform comitia. Much depends on how easily one could travel to Rome for the vote. Until the fourth century, Rome had seventeen tribes, and even on the outskirts of these territories a voter could travel into Rome in well under a day by foot and less than half a day by horse. Voting was no great inconvenience. Beyond that area, however, travel obviously became longer and harder. A city like Tibur, not far from the seventeen ‘old’ tribes, was half a day from Rome even for someone covering 56 km per day on horseback.Footnote 33 The Tiburtines themselves were not Roman citizens, but for voters travelling a similar distance it was presumably impossible to travel to Rome, take the time to vote, and then return home all in one day, especially given the shortened days of winter elections. On horse, from Forum Appii, Ferentinum, Reate, or Narnia, it took a little over a day just to reach Rome. From the tribus Falerna, near Capua, it took two and a half days to reach Rome, and from Ancona, on the Adriatic coast in the tribus Velina, a traveller would need at least four days to reach Rome and the same again to return.
The obstacles to voting were extreme for those at the furthest reaches of Roman territory. Although the seventeen ‘old’ tribes were outnumbered by the eighteen ‘new’ tribes, their constituents must in practice have outvoted the more removed citizens. A simple heuristic table makes the ramifications clear. If we suppose, for ease of arithmetic, a century with 350 voters, drawing 10 voters from each tribe, then the ‘new’ voters would win a majority in each century:
But if we assume that ‘old’ tribesmen were 10 per cent more likely than this baseline to vote and that ‘new’ tribesmen were 10 per cent less likely than the baseline, then the majority swings the other way:
At a 20 per cent shift from the baseline, each election becomes something of a landslide:
In each century, then, the more distant voters were effectively gerrymandered into a minority of each of the one hundred and ninety-three centuries. There was no way that the relatively few voters who made the trip from the faraway Falerna, Teretina, or Velina could avoid being fragmented among the far greater number of voters from closer to Rome. Hall made precisely this point when praising the fairness of the comitia tributa.Footnote 34 The way to prevent such cancellation of distant voices was to do precisely what the reform of 241–218 did: reserve two votes for each tribe, regardless of voter turnout. Now it did not matter if the tribus Fabia had ten times as many voters in attendance as the tribus Velina; each tribe would issue two of the seventy first-class votes in the comitia. Even if only a few voters made the trip from the Adriatic or Campania, the weight of their votes would still reflect the many voters whom they represented, and the ‘new’ tribes would outvote the ‘old’ tribes. This gave greater voting weight to the peripheries, and benefitted rural over urban citizens. The reform brought the voting systems of a local city more in line with the realities of a regional agrarian power.Footnote 35
There was no mathematical precision that ensured that each tribe represented an equal number of voters. Instead, the priority was that each tribe itself had an equal vote. The focus, as Hall showed, was on the ‘group-vote’.Footnote 36 This could become complicated. There were many more tribes close to Rome, so it remained the case that north-west Latium was disproportionately represented. Yet the area around Rome was probably more densely populated than some of the more expansive tribes such as the mountainous Quirina. It was surely the case that some tribes had fewer people than others, and that this increased the personal voting power of those enrolled in the smaller tribes (that is, a relatively small number of people still enjoyed two centuries in the first class). But this was only one side of the coin. The tributum and the dilectus operated through the tribes as well, so while the members of an underpopulated tribe might receive more voting power, they would also have to bear a larger burden to ensure that the tribe was supplying its share of manpower and revenue. This might have balanced the reward and the sacrifice of each tribe's size, or it might be reason to think that censors tried to keep the numbers of the tribes relatively even.
In any case, the focus was on equity not for the individual but for the tribe, the unit that underpinned the reform. This also made practical sense. The demands of the dilectus and of the tributum meant that tribesmen were constantly meeting and coordinating, which made it easy for them to act in a corporate fashion, assessing the views of those interested in the election but unable to make the journey. The associative practices of tribes made it possible for the few who represented the rest at the elections to behave on behalf of all. This might explain why a candidate from the Pollia almost never received a vote from the Papiria; the backstory is that the Pollia voted to condemn the Tusculani in 323, but of interest here is that, even centuries later, the whole Papirian tribe seems to hold this grudge against the whole Pollian tribe (Livy 8.37.12). We may also see such collective tribal decision-making in the elections of 211 (26.22). When the iuniores from the Votinian tribe elected T. Manlius Torquatus, he rejected the honour on the grounds that he was all but blind. The voters at first refused to change their vote, but eventually asked to consult the seniores of their tribe, who suggested other candidates from whom to choose. Livy remarks that young men were eager to consult their wise elders, but this is surely moralizing; the eldest iuniores were in their forties and hence older than many consuls. Moreover, 211 was not so long after the reform was implemented, and it makes no sense that contemporaries who had created a praerogatiua of iuniores doubted that they were capable of voting wisely. Perhaps the voters really did feel that they were representing the whole tribe and wanted to make sure that they and the seniores were in agreement when they all reported back to their home towns.
Whatever the problem that reformers had identified, distribution of centuries by tribe was clearly seen as the key to unlocking it. Mouritsen is correct that the reform reflects an increasing affiliation between Roman citizenship and one's tribal registration; I would take this insight a step further.Footnote 37 The awareness of inequity across territories and the devising of a solution through tribes reflects not only a categorization of people but also a perception of space—and of political space—that was institutionalized in this period.Footnote 38 An official order of the thirty-five tribes was established around this time. As Taylor and Crawford have shown, this order was not by alphabet or by founding date, and was not spatial in the sense of working along cardinal directions; instead, it counted the tribes progressively along each road encountered in an anti-clockwise direction from Romilia, on the way to the coast, all the way back around to the neighbouring Arnensis.Footnote 39 The first four tribes in the official order were the urban ones. The fifth and the sixth were those one would encounter if following the road from Rome to the coast at Ostia (the Via Ostiensis).Footnote 40 Having exhausted the tribes on that road, the person counting tribes then rotated anti-clockwise to the next major road (the Via Appia) and counted the tribes until the end of the line, then rotated again to the next road (the Via Latina) and counted off the tribes encountered until the last was reached, and so on. Once each tribe along each road had been counted across the full 360 degrees, things returned to the Tiber and all thirty-five tribes had been listed. The two most recent tribes (Quirina and Velina) were numbered 29 and 30, so the official order was either overhauled in 241 to prevent them from being tacked on as the last two spots of 34 and 35, or the order itself was only institutionalized after 241. The latter seems more plausible.Footnote 41
Crawford has suggested that the creation of an official order may have sprung from the need to levy emergency troops efficiently to face the Gallic threat of the mid 220s.Footnote 42 It may simply reflect the emergence of a spatial habit of mind in this period, a habit that is also reflected in the contemporaneous reform of the comitia centuriata along tribal—and hence territorial—lines. The list of tribes reflects an ability or a tendency to visualize the tribes serially along roadways that radiate out from the centre (that is, the city) like spokes from a wheel. The viewpoint is that of a person standing on one of the hills of Rome and looking out. Roth has argued that a visualization of Roman Italy displayed in the Temple of Tellus featured a similar hodological depiction of roads radiating from a foregrounded Rome, while Ando has discussed how this perspective can also be seen in the deictic language of Roman law, where terms like ‘here’ and ‘there’ denote what is in Rome and what is in other places.Footnote 43 The conception is also part of an augural worldview and influenced the location of certain sanctuaries.Footnote 44 Though a product of the third century, this habit of mind was not short-lived.Footnote 45 Nearly three centuries later, Q. Veranius, a legate of Claudius’, would employ a clockwise rotation to reel off the roads on a monument in the Lycian city of Patara.Footnote 46
For anyone accustomed to this visualization of Italy as a collection of lands stretching out from Rome, the association of the Roman people with the Roman tribes must have been natural. Thus Clemente could refer to ‘the tribal assembly [as] both the physical manifestation of the people present in Rome at the time of voting, and the result of the voting of all the citizens around Italy’, and that geographical expression of the people now found form in the comitia centuriata as well.Footnote 47 The Commentariolum Petitionis (30) urged the first-century consular candidate to memorize all of Italy tribe by tribe in order to ensure votes in every tribe. The same cultural force that led the Romans to order the tribes serially by distance along roads led them to reorganize centuries in such a way that those tribes nearer to the centre were not overwhelming the votes of those further along the great roadways. They came to visualize the assembling of the Roman people not just as citizens standing in one place but as streaming in along the arterial roads of central Italy. The inequity of having to travel so far along the Via Appia or what would come to be the Via Flaminia must have been obvious to anyone accustomed to view the populus in this hodological way.
The tribal basis of the reform allowed citizens from neighbouring towns—who were bound by camaraderie and by the same systems of taxation and conscription—to vote collectively in the comitia centuriata without overly favouring those who could more easily travel to the city for votes. Tribal leaders may well have been available for consultation in the towns along the main roads as they travelled to Rome. As noted above, however, no explanation is convincing if it explains only part of the reform. The first objection to this territorial theory is that the tribal link was only made in the first class. If associating centuries and tribes was beneficial, why not apply it to all classes? There are two likely reasons. First, descending from the first class to the fifth, there were probably fewer and fewer voters who were going to travel to Rome anyway. If the tribus Velina had centuries reserved for them in the fourth or the fifth class, would there be a single member on site to vote? This would be especially true if the first class was a large one, encompassing around one quarter of all eligible voters.Footnote 48 The more pressing reason, however, is that the proportional mathematics did not allow it. Once the centuries were linked to tribes, they had to be multiples of 35 (presumably 35 or 70). The old Servian comitia, however, had four times as many centuries in the first class (eighty) as in the second, third, or fourth class (twenty each, with thirty in the fifth class). To maintain the timocratic weight in favour of the first class, it was impossible to raise the inferior classes as high as 35 without increasing the first class to 140. This would take the total number of centuries to something approaching three hundred, destroying any semblance to the old number of one hundred and ninety-three centuries in total, which they evidently wished to maintain. To preserve the timocratic calibration of the assembly, only the first class could be linked to the thirty-five tribes, but that was likely the only class with voters from all tribes needing such regional equality.
CONTEXTUALIZING REFORM BETWEEN THE PUNIC WARS
So far I have addressed two of the reforms, arguing that the first class was linked to tribes in order to endow the vote with greater geographical fairness, but that this could not be extended to other classes without granting them more voting power than the timocratic assembly could permit. But this is not enough. Geographically equitable voting explains the ‘tribalizing’ of the centuries of the first class, but the relegation of the sex suffragia, the unequal treatment of different equestrian centuries, and the privilege of the iuniores of becoming the praerogatiua are still unexplained. Nor do the advantages of the tribal redistribution explain why reform happened in this period between the two Punic Wars. I will now argue that the reform becomes comprehensible once we appreciate how the trials of the First Punic War recalibrated the political power of different sections of society. The war produced inequities and failures, resulting in a rejigging of influence in the comitia centuriata.
The First Punic War was, for Polybius (1.63.5), ‘the longest, most continuous and greatest war’. Although the Romans began in spectacular fashion—and did indeed win the war—it was for many years a catalogue of disasters, from M. Atilius Regulus’ failed invasion of Africa to the mind-boggling casualties of serial naval disasters. Loreto has argued that Roman prestige was tarnished by the tortured course of the war.Footnote 49 One consul had been convicted for his sacrilege and incompetence and, as Bleckmann has argued, voters intervened in the management of the war to vent their dissatisfaction as taxpayers, soldiers and sailors.Footnote 50 Never had so much been asked of the Roman people; wealthy voters from the first class, across all the tribes, racked up the sacrifices in money (as tributum) and in men (as casualties).Footnote 51 What is more, following 241, the Romans launched tough campaigns in Gaul and, far from disbanding their expensive fleet, they were eager to deploy two hundred ships to the Adriatic in 229 (Polyb. 2.11.1). Costs would, therefore, continue to run high.
To appreciate the effect of this, we must establish an essential difference between the tribes and the centuries. The tribes were the main vector for civic sacrifice in the Republic. Military service was extracted via a levy (dilectus) based on the tribes, and the burden of taxation (tributum) was apportioned via the tribes. There were privileges associated with tribes—military pay, or voting with respect to laws or lower magistrates through the comitia tributa or concilium plebis—but not in ways that bestowed special honour. It was in the comitia centuriata that the wealthiest had their status recognized. This is where the magistrates with the highest prestige were elected. The unreformed Servian comitia centuriata thus embodied one side of a civic imbalance. Sacrifice was distributed tribally in a way that equitably touched upon every farm in every territory, but the privilege of rank was much less evenly spread; even for those of equal wealth, it was less and less attainable as one moved away from north-west Latium or southern Etruria. As the scale of demands grew in each tribe, the geographical inequities became less acceptable. Why should the privileges of rich men in Campania face such obstacles, given that, through the tribus Falerna, they were paying the same steep taxes and they were sacrificing their own beloved sons and clients?Footnote 52 The reforms of 241–218 granted them the equal representation that their equal sacrifice demanded. Such measures would be even more necessary with the addition of the Velina and the Quirina. In the context of the First Punic War's crushing demands and assertive voters, reform makes a good deal of sense, ensuring that privilege was distributed in the same way in which burdens were distributed.
But why would this result in the division of young and old? It is undeniable that the vote of the seniores benefitted from this bifurcation by age; younger Romans outnumbered older Romans, yet each demographic received equal centuries, which must have enhanced the voting power of the elderly. The voting power of relatively few seniores matched the voting power of relatively many iuniores. Yet the iuniores received the privilege of voting first. This may reflect the discordant place of seniores in a notional military assembly. The comitia was constructed as the people under arms, yet the seniores were defined by the fact that they no longer took up arms; within the tribes, the iunior/senior division established who was eligible for (emergency) conscription and who was not. The older men no doubt had legitimate grievances in terms of the taxation they had paid and the sacrifices that their communities endured in war, but, if the character of the comitia centuriata was to be preserved, the voice of the young men who were liable to serve had to be elevated. Isolating a vote for the iuniores, and elevating it through the praerogatiua, preserved the essential military nature of the comitia centuriata. Well might the older voters complain about the tributum they were paying and demand a weightier voice as compensation, but they were not the ones who would live or die according to the competence of the consul; after all, the point of the comitia centuriata was to ensure that those who might serve in the upcoming summer campaigns were able to choose the commander who would lead them. Isolating the iuniores and granting them the first vote reinforced that.
Moreover, if the reform dated to the years immediately after the First Punic War—something unfortunately unknowable—the younger men may have felt that the old veterans of the Italian Wars knew nothing of the new landscape of naval war and overseas campaigns. The term sexagenarius was shorthand for any old man, and it is thus dangerous to read too much into the specifics of any precise age cohort.Footnote 53 Yet the arithmetic is tantalizing. Any sexagenarius in 238 or 237 would have escaped even the initial years of the First Punic War. Who were these old sexagenarii to decide who should lead a fleet? They knew nothing of this new warfare, and it was their votes that had elected the men who lost hundreds of ships.Footnote 54 In any case, the interests and demands of older taxpayers were reflected in the creation of centuries for the seniores, but, if the raison d’être of the military comitia was to be upheld, then the most important voice—that of the comitia praerogatiua—had to be reserved for the active soldiers and for the most recent veterans who knew the landscape of contemporary wars and had experience of the electoral candidates as military tribunes and quaestors.Footnote 55 Mouritsen's argument that the centuria praerogatiua was not a feature of legislative votes, trials, or declarations of war becomes especially revealing;Footnote 56 if the iuniores did not enjoy an instituted first vote in these other circumstances, then it is even clearer that their privilege was specifically—even solely—intended to give the troops the lead in choosing their own commander.
This is especially meaningful given that so many consuls had performed so poorly in the second half of the First Punic War, which brings us to the demotion of the sex suffragia. A striking feature of the war is that it never became synonymous with any figure's brilliance in the way in which the Great Samnite War was associated with Q. Fabius Rullianus and P. Decius Mus, or the Second Punic War elevated Q. Fabius Cunctator, M. Claudius Marcellus and P. Cornelius Scipio Africanus. C. Duillius earned a famous victory and commensurate honours, but he never earned a second consulship.Footnote 57 C. Lutatius Catulus ended the war, but his peace terms were overruled by the voters and he had little impact on the two decades that preceded his consulship. L. Metellus dominated the middle years of the conflict, but never earned a decisive victory to conclude matters. Other famous figures to emerge from the First Punic War were M. Atilius Regulus, whose virtue was in the context of defeat, and the notorious P. Claudius Pulcher, whose disregard of the sacred chickens cost so many lives and ships. Though victory was snatched in the end, repeated failures meant that the prestige of the political class was dulled by 241. It is in this context that the relegation of the sex suffragia ought to be understood.
Precisely who was enrolled in the sex suffragia—and how they differed from the other twelve centuries of equites—is not explicitly stated, but it is all but certain that the senators were among them.Footnote 58 Stone has argued that the six centuries included the three hundred senators as well as three hundred of the most elite ex-cavalrymen ‘domiciled in Rome’. Though they decided against careers in politics, these distinguished men were enrolled by the censors on the same optimus quisque principle that the Lex Ouinia established for selecting senators.Footnote 59 The other twelve centuries, according to this theory, comprised the rest of the equites equo publico (rich men honoured with a publicly funded horse, as opposed to rich men who fought on their own horses and voted in the first class), who tended to live further from the city. If the voters in the sex suffragia were indeed located near Rome, this itself would be reason to dislocate them from a reform that sought to equalize the votes of tribes on a geographically fair basis; indeed, the most elite voters in more distant municipia presumably belonged to, or aspired to belong to, the remaining twelve centuries of equites. Even if Stone's ‘hypothetical template’ is not accurate in every detail, however, it must be true that the sex suffragia contained the most illustrious Romans—including senators—and that old Roman families dominated their six hundred places. These were the men who had led the voting during the war and, in the Senate, had run its overall strategy. Their demotion—along with the creation of a centuria praerogatiua of iuniores—can be interpreted not simply as a push to equalize Rome's geographical centre and periphery but also as an expression of the soldiery's dissatisfaction with the leadership of the 250s and the 240s.
To tie all this together is to situate reform in the wake of the ordeal of the First Punic War and what may have been a lingering perception that deep civic sacrifice had been betrayed by substandard leadership. Voters would justly have had two related complaints in this. The first was that, although sacrifice was distributed evenly across the Roman citizen body via the tribes, when it came to electing consuls, the centuries disproportionately advantaged those closer to Rome; this was unacceptable to those more distant Romans who had sacrificed so much. To make matters worse, the centrally located Romans who had dominated voting had voted for consuls of a lower quality than might have been desired. It would make sense, then, that there were demands for greater voting power on the part of those who lived further from Rome, and this would result in a geographical distribution of centuries. Reform appears perfectly designed to address the complaints of these wealthy men that their sacrifice in sons and taxes were not reflected in voting power. At the same time, through division of the age groups, the men of conscription age put their symbolic and practical stamp on the assembly that elected their commanders. As the system was being redrawn, they were given the right to vote first. The result was a recalibration of voting power that accorded both with the venerable sense that the comitia centuriata was the citizenry under arms, and with a new sense of whose contributions and sacrifices entitled them to what degree of political prominence. All tribes received equal representation in the ever-important first class now, just as all were subject to tributum and the dilectus; but the reformers never ceased to be aware that the young men subject to conscription—and the recent veterans who had witnessed how the consular candidates had performed as junior magistrates—had to have the leading voice in who would become consuls and lead the next year's armies.
CONCLUSIONS
The reform of the comitia centuriata between the First and the Second Punic Wars reduced the voting power of the richest Romans, and it stood to reason that some scholars would see this as democratic. The changes were not enough, however, to sustain such claims of democratization, and so they were dismissed by other scholars. As so often, much comes down to definitions. In the Aristotelian sense, this was surely not democratic; it did nothing to aid the poor and barely dented the lock that the wealthy enjoyed on the vote. But it was designed to increase fairness in other ways. Though the reform would not upset the distribution of voting power between rich and poor, it did change the distribution of voting power between rich men who lived close to Rome and rich men who lived further away. Furthermore, it tilted the focus away from the narrow elite that belonged to the sex suffragia towards the broader class of wealthy landowners who dominated the first class and the cavalry. It increased voter equity within the broader oligarchic regime that dominated the comitia centuriata by advancing the interests of municipal elites in the first class. Taylor's suggestion that the driving force behind reform was the class of wealthy men who fought as equites equo suo but were not enrolled in the equestrian centuries is tempting here; they voted in the first class, and they successfully pushed those centuries to the fore at the expense of ‘the more aristocratic (though not necessarily richer) equites with public horse’.Footnote 60 It might also be tempting to hypothesize a role for C. Flaminius here; his electoral record suggests a strong relationship with the comitia centuriata, though if such a figure were associated with reform, we might have expected to hear something of it.Footnote 61
Inclusion across regions became a feature of the assembly. Many of the beneficiaries lived in the many municipia that had fought against Rome in the past century or so, and thus we can see this as a way for both parties to negotiate the transition from hostility to cohesion. The unusual extension of citizenship to people who had never previously been Roman meant that a domestic issue such as voting reform could occupy something akin to diplomacy or foreign relations. There is no way to know how many people in the far-flung tribes were exported Roman settlers and how many were newly enfranchised non-Romans, but either way voting reform was a way to reach out to them, to bind them to the res publica, and to cement their participation and loyalty. It thus constituted both domestic Roman politics and wider intra-Italian politics. At the same time, an insistence on the part of the municipal elites that they wanted a role in a Roman assembly was itself a gesture that the past was passed, and that they accepted and desired a part in Rome's polity. If one were to accept Terrenato's vision of Rome as the federal capital of Central Italy, this is precisely the sort of vision of regional representation that one might expect.Footnote 62 Or it might reflect a different process, one that exemplified a more centralized, institutional and affective force of integration.
The polity could be under no illusions by 241 or during the Gallic invasions of the 220s about whether they were all in this together. They shared costs and sacrifices, as well as victories. The blood of Latins, Campanians, Sabines and Picentes all mixed on the fields where so many fought together and in the seas where so many ships sank. The new comitia centuriata ensured that wealthy men from all across Roman territory would be able to elect consuls with a degree of equal representation that reflected their equal sacrifice. But beyond elections the First Punic War may well have been punctuated by other consequential votes as well. The war itself may have begun with a vote to aid the Mamertines.Footnote 63 If Zonaras is to be believed, there were votes that intervened in the conduct of war in 253 and after a pair of calamities in 249, while voters rejected the initial peace terms negotiated with the Carthaginians (though whether these were all in the comitia centuriata is debatable at best).Footnote 64 It is possible that the real target of voting reform was not the elections but these votes on policy and on declarations of war, with voters clamouring for a say in whether and how wars were waged. When such matters arose, moreover, it seems that the iuniores lost their privilege of the centuria praerogatiua, and the views of the seniores could consequently be elevated. But every rich landowner could feel satisfied that his voice was counted in a century now, no matter where he lived. This may have cost the political elite some degree of control, but to the extent that it increased the cohesion of wealthy Romans from coast to coast by the time that Hannibal arrived, it was well worth the investment.
Open access
