Jump to ContentJump to Main Navigation
Registration and RecognitionDocumenting the Person in World History$

Keith Breckenridge and Simon Szreter

Print publication date: 2012

Print ISBN-13: 9780197265314

Published to British Academy Scholarship Online: January 2014

DOI: 10.5871/bacad/9780197265314.001.0001

Show Summary Details

Identity Registration in the Classical Mediterranean World

Identity Registration in the Classical Mediterranean World

(p.169) 6 Identity Registration in the Classical Mediterranean World
Registration and Recognition

Rebecca Flemming

British Academy

Abstract and Keywords

This chapter provides an analytical survey of identity registration practices (and their absence) across the ancient Mediterranean. The discussion focuses on the citizen census in Republican Rome, but also covers registration activities in democratic Athens, with further reference to a wider background including Ptolemaic Egypt and the Roman Empire. The analysis emphasizes that civil registration in ancient states is more a matter of record and ritual than actual legal definition, and that its systems are as much about membership and privilege, as about subordination and exploitation.

Keywords:   ancient history, social hierarchy, state formation, citizen census, identity registration practices

THE SECOND CHAPTER OF LUKE’S GOSPEL famously opens with the statement:

Now it happened in those days that a decree went out from Ceasar Augustus that all the world should be registered (apographesthai). This registration (apographê) first occurred when Quirinus was governor of Syria. And all went to be registered, each to his own city.

(Luke 2: 1–3)

Other sources confirm that, around AD 6, Judaea (up until then a ‘client kingdom’ of Rome, under the rule of Herod the Great and his, rather less great, successors) became a prefecture under the overall control of the Roman governor of Syria, an annexe of that province, and, as part of this provincial reorganization a census was conducted, by Quirinus and his officials.1

A provincial census, an overall assessment of the wealth and human resources of conquered territory, was a standard part of the empire-building process under Augustus. Victor in the final episodes of the civil wars which eventually pulled Rome’s republican system apart (most decisively victorious in the battle of Actium in 31 BC), Augustus had to work hard to re-establish a stable political system in Rome and her empire, a regime now centred around himself as the ‘foremost/first man’ (princeps) of the Roman state, and which he would pass on to his selected successors on his death in AD 14. Extensive overseas conquests, and the consolidation of those gains, while developing more integrated structures of imperial rule across all of Rome’s vast domains, were a key part of this project. A provincial census of the Gauls in 27 BC was followed by similar surveys of people and their assets, not just in Syria, but also, for example, in Lusitania, and probably (p.170) Egypt.2 That the purpose of these registers was essentially exploitative – mainly to calculate the amount of tribute that would be paid by a province, by its various cities and districts, in taxation to Rome, on occasion also to enable the levying of troops to serve in Rome’s armies – is clear, and the first census might be met by local resistance, even revolt, before the system settled down.3

In some provinces, however, the local population was already familiar with a system of registration for the purposes of exploitation. Under its previous rulers – the Macedonian Ptolemies – Egypt had developed a complex and encompassing fiscal and registrative package.4 They, too, were building on past, Pharaonic (and probably also Persian), practice, with catalogues of cattle dating back to the Old Kingdom (c. 2400 BC) and of people (and animals) back to the early New Kingdom (c. 1450 BC). These latter catalogues were especially concerned to list occupations since the regime relied heavily on corvée labour for a range of particular royal projects, and for the general maintenance of the irrigation system on which Egyptian agriculture depended. Under the Ptolemies the collection of information about the population became focused on taxation – on the extraction of tax revenues, monetary revenues, centred on the (almost) universal ‘salt tax’ (halikê) – and the process became more literate, with written documents – household declarations, tax-registers, and receipts – proliferating.5 An already substantial bureaucracy was adapted to the new demands, the changing shape and operation, of Macedonian rule.

Still, though the slippage involved in the Authorized Version’s translation of Luke, in which Augustus decreed that ‘all the world should be taxed’, is perhaps acceptable in an imperial context, it is misleading in many others. ‘Registration’ had quite different valences and purposes in other polities in the classical Mediterranean world, including, for example, democratic Athens in the fifth and fourth centuries BC, and, indeed, within the citizen census of the Roman Republic, on which later provincial practice was partly based. This latter will be the main focus of this essay, as the most richly attested of ancient registration cultures, and the one which best displays the complexities involved; but it is also worth saying a bit more about Athens, both as a valuable counterpoint to Republican Rome in (p.171) particular, and as further illustration of the variety of ways in which ancient states explicitly recognized their people, and recorded that recognition.

For, while by no means identical kinds of states, democratic Athens (c.503–338 BC) and Republican Rome (c. 509–31 BC) shared certain key features which contrast, to some extent, with the monarchic and imperial formations discussed so far. Both were, at least in their origins, autonomous city-states, constructed around concepts of shared community membership – citizenship – and collective self-governance. Interactions between the state and the people were, in this context, self-encounters, rather than crossing any real divide, a point emphasized by the lack of any developed state apparatus in either case. Similarly, the armies of both democratic Athens and Republican Rome were, essentially, citizen militia, the citizenry under arms, and military service was as much a privilege as an obligation of belonging to the community, heavily valorized in Greek and Roman culture. Moreover, the two were resistant to imposing any form of regular direct taxation on their citizens, an ideological opposition that found practical support in the resources of empire in both cases, and also for Athens, the silver mines at nearby Laureion. The intrusion of empire, again, into this narrative serves, of course, to emphasize that the entitlements of membership may come at the expense of others: there are always exclusions, the subordination and subjection of non-citizens. Their arrangement of these latter elements is part of what differs between Athens and Rome, and is reflected in their divergent approaches to civil registration, but this divergence extends well beyond such variation, too, responding to a whole range of factors, in interesting, and perhaps even unpredictable, ways.

Democratic Athens

Democratic Athens was not much interested in the enumeration of its people, but it did systematically register sections of the population, albeit in quite a distinctive way. Two groups went on the most official lists – that is adult male citizens, on the one hand, and adult male and some adult female metics (metoikoi, ‘resident aliens’) and manumitted slaves, on the other – groups which occupied quite different positions in Athenian society, one politically pre-eminent, the other marginal; and which were recorded for contrasting reasons.6 Citizen registration publicly recognized (p.172) that a man had now become a fully participating member of the democracy; it affirmed that pre-eminence, within a specific institutional framework, and while the registration of foreign residents and freed slaves partook of that same frame-work, it functioned to establish a relation of subordination within the workings of the polis, to confirm that marginality. There were other lists around – relating to aspects of military service, for instance, and (probably) various indirect taxes – but they lack even the somewhat fragile significance of the citizen and metic registers.

An aspect of this fragility, the distinctive character of Athenian civic registration, was its local organization. This had its origins in the reforms of Cleisthenes in 508/507 BC, which divided the city and its surrounding territory into ‘demes’ and made them the basis of the new political system which developed: that is, democracy.7 The matrix through which citizenship was now to be recognized and organized was this network of demes, geographically defined political entities which were carefully grouped into ten tribes (phylai), to provide the basic political infrastructure for democratic Athens. Exactly how many demes Cleisthenes established is unclear, but the figure is unlikely to be very different from the 139 of the classical polis; a city-state encompassing a territory of around 2,500 square kilometres, and a citizen population in the region of 100,000 (men, women and children).8 As constituent wards of the city, and towns, villages and hamlets of the surrounding territory of Attica, the demes varied in size (though the numbers they contained were balanced out at the tribal level), but none would have been that large, as they were designed to comprise units of communal knowledge and control. Each had a leading official – a ‘demarch’ – and an assembly, among other organizational features, and these were the main players in the processes of registration for both citizens and metics. While institutions of the polis itself, the council and courts, had some oversight of, and engagement with, these activities, there is scant evidence for any central compilation of deme lists: there was no universal register of citizens, and perhaps no central register of anyone else either.

Composed around the middle of the fourth century BC, the Aristotelian work on the ‘constitution’ (politeia) of the Athenians describes the then-current process of citizen registration in some detail:

Men born of citizen (astoi) parentage on both sides belong to the political/citizen-body (politeia), they are registered (engraphesthai) among the demesmen when they have reached the age of eighteen. When they are being enrolled, the demesmen decide, by vote under oath, first, whether they appear to have reached the age which the law (p.173) prescribes – and, if they do not so appear, they return to being ‘boys’ once again – and secondly, whether [each one] is free and born in accordance with the laws. Then, if they should vote against him as being not free, he appeals to the jury-court (dikastêrion), and the demesmen choose five of their number as accusers; if it is decided [by the court] that he has no right to be registered, the polis sells him [into slavery], but if he wins the demesmen are obliged to register him. After this the council (boulê) scrutinises those who have been registered, and if it is decided that someone is younger than eighteen, the demesmen who registered him are fined.9

There is some debate about the accuracy of this account, even for the fourth century: in particular, the strict division of jurisdiction in respect to decisions about age, and about parentage, seems unlikely in practice; but the basic picture is clear enough. On reaching the age of 18, a young man born of parents who are both citizens (astoi: belong to the city, the astu) is registered in the deme, that is, he is written into the list of demesmen – the lexiarchikon grammateion – which appears elsewhere in the historical record. This registration occurs in a deme assembly, and all the demesmen present vote (under oath, as always) on whether those before them have reached the required age, and are the free and legitimate sons of citizen parents, that is, were born in accordance with Pericles’ citizenship law of 451/450 BC, which required a citizen mother as well as a citizen father.10 This vote is appealable in the city’s law-courts, though the stakes are high, and the council of the city also scrutinizes those newly registered. The same deme assembly, probably held near the beginning of the administrative year, would also vote to enrol adoptive rather than natural sons of demesmen, and any newly enfranchised aliens who had selected, or been allotted to, that deme.11 Since the requirements for adoption were the same as for citizenship, that is to say only legitimate children of two astoi could be adopted, and the actual enfranchisement of foreigners was effected by a vote of the assembly of the whole city, in all these cases therefore the deme registers those who are already citizens, by birth, adoption, or decree. This process of registration does not make citizens, but officially recognizes them, and records that recognition, within the public political setting of the deme.

The sense in which registration is a matter of ritual and record but not definition is emphasized in a range of contexts: most obviously in legal disputes about citizenship involving the deme-lists themselves. A general scrutiny of all the local registers ordered in 346/345 BC, for instance, led to some men being struck off and (p.174) then fighting for reinstatement in the courts.12 Allegations of personal enmity, manipulation of the deme assembly, and false accusations are combined in such cases with positive claims of status based on a network of living witnesses to key communal events which speak to the individual’s legitimate citizen parentage, and broader civic participation.13 All this is more important than being on a list or not, and there is little evidence that the lexiarchika grammateia were ever used to check eligibility for the various entitlements that being ‘of the politeia’ brought with it in democratic Athens: that is, participation in the assembly, council, and law-courts, and the holding of public office. Moreover, though Pericles’ citizenship law brought women’s status as astoi (or not) into sharper relief, they never appeared on any register.14 The absence of any reference to the deme-lists in connection with eligibility for public office is particularly striking, since there was a process of scrutiny for each candidate, in which use of deme registers would certainly have been very practical, but in which personal testimony is systematically preferred.

Beyond citizens themselves, metics and freed-persons were also registered in their demes of residence: when their stay became long enough to require it, following manumission, or, presumably, on attaining adulthood, or just moving house, within these categories. They also had to register a ‘prostates’, an Athenian citizen who acted as some kind of legal sponsor or representative, who provided a point of reference for them in the Athenian legal and civic system.15 On registering they also became liable for military service and direct taxation. The ‘metoikion’, the metic-tax, which was, in many ways, the most important marker of their subordinate status, was a kind of taxation that no Athenian citizen was burdened with, levied at the rate of twelve drachmae a year for a man and six drachmae for a woman (that is, an independent woman who had neither a husband nor a son to pay for her); the relationship of manumitted slaves to this tax is unclear, but they do seem to have had to make some payments.16 The metoikion was farmed out by the board of officials who organized state contracts (the poletai) to tax collectors on an annual basis. How this all operated in practice is uncertain, however, and though farming out a fixed tax suggests that the central authorities, the poletai and (p.175) the council under whose auspices they worked, knew how many metics there were, information which would have been most easily acquired through regular collation of deme registers, there is no evidence that this process in fact occurred.17 Even the figure of 10,000 metics produced in the population count organized by Demetrius of Phalerum, after he took control of Athens on behalf of their Macedonian overlords in 317 BC,looks more like an estimate of the kind that might be used for such tax purposes than a careful calculation based on information from the demes.18 It does provide a sense of the rough size of this (substantial) group, however, or, at least of its adult, tax-paying portion.

In many ways, however, it was the non-citizens who paid the price for this rather casual attitude to registration, and, indeed, for the non-authoritative nature of Athenian registers in general. Both resident aliens and freed slaves were permanently at risk of being charged with not having a prostates, and not paying their taxes, and if convicted the penalty was enslavement. The precise nature of the action brought against them is debated, and this seems to be an area in which the foreign resident and former slave diverge significantly, but such cases have certainly made their mark on the historical record in various ways, testifying to the vulnerable position of non-citizens in Athens.19 A good defence, or indeed deterrent, would have been provided by a more robust public registration system. If official lists of metics, their prostatai, and tax payments had been accessible and authoritative, then the sorts of events described, for example, in a Demosthenic law-court speech from the fourth century BC would have been avoided.20 The case is against Aristogeiton, for state debt, and as part of the customary demolition of character it is alleged that he seized his metic mistress Zobia (who had done nothing but help him) and dragged her off to the office of the tax collectors, where, if it had not happened that her metoikion had been paid, she would have been sold. The Greek is unclear about the means by which it is established that she was not in arrears, whether this was a matter of written record or not; but certainly, in order to support these accusations against Aristogeiton, Zobia’s prostates is called as a witness, and the tax collectors: no documents are adduced, or referred to.21

(p.176) Registration in democratic Athens was, therefore, distinctive in many ways, particularly in its local organization, and its performative character. Registration was a ritual, a social event, rather than a process which generated an authoritative document, a ‘register’ in any very strong sense of the term. Adult male citizens fully participated in that ritual, both had their status validated and validated the status of others; while metics’ involvement was more partial, their agency compromised by their registration with a prostates, and their concomitant inability to witness others’ status: the ritual was one of subordinated integration. This fitted in with, enacted, the key values, and organizational principles, of the democracy, which was based on public participation, personal and family networks, and an exclusive sense of community.

Republican Rome

In contrast to the rather patchy registration practices of classical Greek city-states, such as that outlined for democratic Athens, and their lack of interest in enumeration, Rome adopted a systematic approach to both. The Roman ‘census’, and the word is theirs, combined record, ordering, moral scrutiny, and counting of their citizen population, and involved interaction with the divine. It began, so the stories go, back in the time when the city was ruled by kings and was in the process of formation as community and polity; it continued into the Republic, traditionally founded in 509 BC, when elected magistrates replaced the expelled King Tarqin, and the informal royal council and popular assembly started to move towards creating the powerful political partnership of ‘senate and people of Rome’; and it managed the transition from central Italian city-state to global empire over the centuries to come. There were changes along the way, most significant perhaps in the aftermath of the Social War (91–89 BC), when Rome granted citizenship to all the communities in Italy south of the Po, and became a quite different kind of state than previously – though not a very stable one, as the Republic would last only about half a century longer and Roman governance would once again take a more monarchic form under Augustus, after 31 BC.

It was not only Roman power and territory, but also people, the Roman citizenry – cives Romani – that increased over this period (with some losses along the way), and not just in lock-step with the other two. The most decisive episode of enlargement was the enfranchisement of Italy during and after the Social War (fought against some of Rome’s Italian allies, that is her ‘socii’); and Roman notions of citizenship were always more inclusive than the Greek. The citizen-body, the ‘civitas’, was steadily augmented in the Republic by the manumission of slaves, immigration to the metropolis (and some Roman colonies), and a range of grants to individuals and groups; as well as some early territorial incorporations, and later colonial foundations. It is particularly worth noting that in Rome the freed (p.177) slave became a citizen, a slightly hampered citizen in the first instance, but with all legal impediments removed in a generation, in contrast to the Greek tradition, already described for Athens, in which manumitted slaves joined the ranks of the free non-citizens; but this is part of a more generally open approach of the Romans to civitas. Not that the borders of the category were unguarded, or that status did not matter. Roman citizenship brought with it a rich set of entitlements and obligations, political and military, legal and religious, social, economic, and ideological, with women being excluded from the first two domains, but certainly not the rest – the wider patterns of communal membership and participation; though the expansion of citizenship shifted the balance between these elements, and would eventually diffuse its importance.22

The figures recorded in the annals of Roman history give a sense of this expanding citizen body, despite the many problems that attend on them. So, a census and ‘lustrum’ – the latter being the complex sacrifice and communal ritual which completed the assessment process – are reported to have been conducted in 508 BC, in Rome’s first year as a Republic, with the number of assessed citizens announced as 130,000.23 In the lustrum of 234/233 BC, during the decades of Rome’s Italian hegemony prior to Hannibal’s invasion of the peninsula, and when, according to Peter Brunt in his magisterial study Italian Manpower: 225 BC–AD 14 ([1971]1987), accounts begin to be more reliable, the declaration was 270,713 assessed citizens, a number which had risen to 900,000 in the lustrum of 70/69 BC, not long after the Social War.24 This was the final, full census of the Republic, when the system was already unravelling; but a return to traditional practice in this area was a key part of Augustus’ programme of political and cultural restoration after the civil wars. As he himself proclaims: ‘I conducted a census of the people with Marcus Agrippa as my colleague. I performed the lustrum 42 years after the last one [i.e. in 28 BC]; in that lustrum the assessed heads of Roman citizens were 4,063,000’ (Res Gestae 8.2).

Debate currently rages about the accuracy of these figures and what they represent.25 Are these the numbers of adult male citizens, as in Athens, for example; or all citizens – men, women and children – as were certainly recorded later in Roman Egypt, for instance; or was the calculation according to some other principle? Was the procedure for the census stable across the centuries, but perhaps more or less rigorously carried out, or did the methods change, in particular to achieve the massive Augustan figure in 28 BC? These are all important questions, which have (p.178) a dramatic impact on how the demography of Roman Italy is understood, with wider ramifications for the history of the Roman Republic; but the focus here is on the process and socio-cultural significance of registration not the statistical results of enumeration, hitherto a rather overlooked part of the whole business, but itself crucial in many respects. Still, a rough sense of the scale of the undertaking is vital, as also a feel for the level of Roman self-consciousness about their population practices. That censuses and lustra were a standard feature of Roman historical narratives and formed part of Rome’s communal record demonstrates their contemporary significance, a point which Augustus’ actions in this respect emphasize further.26 One highly influential Roman historian, Livy, writing in the Augustan period, even goes so far as to assert that Rome’s rise from insignificant central Italian city to world superpower was predicated, in no small part, on its regular assessment of the citizen-body: the census was, for him, one of the most fundamental, and beneficial, institutions of the Roman state.27 Not, it should be said, because it produced useful data, but because it engendered and sustained good order and effective organization in war and peace; indeed, the number with which Livy concludes his account of the first ever census and lustrum in Rome comes as something of a surprise, as counting the citizens had not been mentioned in the discussion up till then, even though it would become integral to the process.28

In terms of regularity, Republican censuses occurred on a roughly quinquennial schedule, with various gaps, crises, and procedural developments along the way.29 From 443 BC, two ‘censors’ were elected to see each assessment cycle through, which was expected to take around eighteen months, as well as carrying out a growing number of other tasks and responsibilities, which might take longer. The office was one of great prestige and some power, which usually came towards the end of an individual’s sequence of magistracies, and so was held by senior and distinguished figures from the Roman elite. Once the censors opened the census, it was the duty of Roman citizens to present themselves before the competent authorities, and to provide the information required under the formula censoria, under oath. They had to declare, orally, their full name, their father’s name (or their patron’s name if a freedman; that is, the name of the former master who had manumitted them), their tribe, their age, and their property, giving both a summary description and a valuation. They may also have had to supply the names and ages of their children, and perhaps the name of their wife also; certainly they were often (p.179) asked about their marital and generative situation.30 Further undertakings about their conduct might also be demanded, as the censors saw fit. The declaration of information – professio – was recorded in writing, by scribes, and a complete account for each census was compiled, which could be consulted in certain circumstances, as will be discussed later.

The censors had extensive powers of enforcement and judgement. A fundamental aspect of the census was the ordering of the people into six ‘classes’, sub-divided into centuries, on a scale of property qualifications, and their distribution into the thirty-five Roman ‘tribes’, with both allocations essentially under the control of the censors. The level of the property qualifications had already been established, and, though originally geographical entities, tribes were now hereditary, so the situation should have been made clear in the declaration, the rough accuracy of which was generally assured by its nature as a public, oral oath, subject to censorial scrutiny. However, the censors could change the rules, counting possessions they disapproved of as valued differently from land, for example, or moving politically problematic groups – such as freedmen – between tribes, more or less as they wished.31 Other, more individual, punitive options were also available to them, and might be deployed as punishment for a range of misdemeanours, from evasion of military service to neglect of agricultural land and mocking the censorial process itself; but might also result from simple personal animosity.32

Alongside its ideological significance through enactment of general Roman principles of hierarchy and order, this ranking and placement mattered practically in a number of ways. Initially focused on military organization – on mustering, structuring, and paying for the army – at least as the stories go, as the Roman Republican state developed it was the political ramifications of the census which became most crucial. Still, until the reforms of Marius in 106 BC, those in the lowest class were excluded from serving in the legions, though they might (and did) serve in the navy; and both tribe and century played some role in the organization of the military levy (dilectus).33 Similarly, until Roman victories in the wealthy Hellenistic (p.180) East in 167 BC, liability for tributum – the closest Rome came to directly taxing its citizens – was determined by class: the wealthier contributed most to the cost of any campaign, and the sixth class, the poorest, were again pretty much exempt.34 But it was in terms of power that the real pay-back for registering and being assessed occurred, at least for those with plenty of property to declare. In the Republican political system, popular assemblies, organized by both century and tribe, were the locus of annual elections for the various magistracies and of voting on legislative proposals, war and peace, and other matters of state. Because of the particular form of group voting which characterized these assemblies – in which all groups counted equally, regardless of size, and votes were cast in a certain order – being in the top centuries gave an individual’s vote far greater weight. There were eighty ordinary centuries in the prima classis, the ‘first class’, together with a further eighteen special centuries of cavalrymen (equites), in comparison to twenty or thirty for the next four (alongside a few extra), and just one at the bottom. Also, being in one of the thirty-one smaller, more exclusive, ‘rural’ tribes (and not in one of the four huge and sprawling ‘urban’ ones) increased the value of an individual’s vote. Indeed, with 193 centuries in total, many elections and issues were decided entirely within the top ranks. In addition, only those in the prima classis were eligible to stand for election.

What the censors could not do, however, was expel anyone from the system as a whole. C. Claudius Pulcher, censor in 169 BC, states that: ‘to take away citizenship (civitas) and freedom (libertas); not simply to determine where a person will be assessed (censere), but to exclude them from the census entirely’, is outside his powers (Livy 45.15.1–6). Nor could they, or anyone else, compel participation in the census. It is reported that Servius Tullius, the early king credited with invention of the Roman census system, instituted severe penalties for non-registration: the threat of death or imprisonment, according to Livy (1.44.1), or the forfeit of property, being whipped and sold into slavery, according to another author of the Augustan period.35 However, this does not seem to have been a Republican reality, though deceitful, ill-intentioned, avoidance of the census may have been actionable and punishable (by flogging and sale of property).36 No such cases, or penalties, (p.181) in relation to incensi – those who have not given an account of themselves and their property to the authorities – are recorded, despite the occasional loose evocation of such possibilities, and plenty of general reportage around the census, including punishments for other things. Indeed, instances of failure, or inability, to appear before the censors proliferate over the course of the Republic. Troops serving overseas at the relevant time were not generally included, for example: Livy mentions that, in 204 BC, the lustrum was unusually delayed because the censors had sent out across the provinces to discover how many Roman citizens were stationed in the various armies around the empire (29.37.6). A more individual case is described by Cicero, the great statesman and orator of the late Republic: L. Annius was able to avoid the strictures of the lex Voconia – a law preventing those in the prima classis from instituting female heirs – because, though certainly wealthy enough to be affected, he was incensus, and so had not been assessed; and Cicero does not condemn him for that state, nor does he suggest that anyone else would either.37

The point is, as all the versions of the Servian penalties underline (just as they emphasize the intrinsically coercive nature of a royal state), that citizenship did not depend on the census and lustrum. It was generated elsewhere, by birth and law, just as in Athens. Communal membership and all that entailed had a separate existence, which the census could record, but did not engender, or even prove. This is explicitly stated by Cicero in a famous law case of 62 BC: ‘The census does not establish citizenship (ius civitatis)’, he unequivocally asserts (Pro Archia 11). And though he is defending a man – Archias – whose citizen status has been challenged, and who had not been included in any Roman census, and so is hardly going to argue the reverse, this fits so well with all the other evidence that there is little reason to doubt him. Everybody knows, says Cicero, that his client was away from Rome, on campaign with the general Lucullus, at the times of the last two censuses. Moreover, Cicero can offer a mixture of personal testimony and written record to demonstrate that Archias (originally a Greek from Antioch) followed proper procedure in becoming a Roman citizen, and that he lived in Rome as a citizen, doing all the things good Roman citizens do, such as making wills and receiving legacies under Roman law, and participating in various patterns of public reward and recognition. As in classical Athens, the emphasis is on living, respectable and reliable, witnesses to the various stages of the process of acquiring civitas; but where official documents – publicae tabulae – are available, Cicero certainly refers to them – they do have some weight.38

(p.182) The sense in which documentation, its absence and presence, was part of the legal debate, on both sides, is significant, however, despite its lack of decisiveness.39 It is clear that the products of various kinds of registration could be referred to, in a meaningful way, and it has been suggested – mostly by those with enumerative interests in the census as a reasonably complete account – that being census, being on the record, was of practical use to those at the bottom, as well as the top, of Roman society; that registration was generally beneficial to all individual citizens as well as the state, with its organizational and ideological concerns. Thus, Brunt argues that being on the register would have helped even the lowest down the social order assert the legal privileges and voting rights of the civis Romanus.40 These claims are hard to support, however, since, though Roman voting procedure did allow for individual scrutiny of each voter, there is no evidence that, at least in the tribal assemblies, this was done on anything other than a very informal level (if at all); and what went on in the centuriate assembly, where the top centuries were small enough to be self-regulating, was irrelevant to the majority.41 Moreover, as Archias demonstrates, access to the Roman legal system was not particularly assisted by being in the census. However, his case did not involve perhaps the most paradigmatic of Roman citizen rights, that of provocatio – the (theoretical) guarantee against execution without trial, and, by the middle of the Republic, also against flogging – which Brunt certainly draws attention to in this context.42 However, in most of the recorded examples where provocatio is invoked, status is not at issue; these cases are all about the proper and improper exercise of power more broadly. The one exception to this rule – though rather later (and, again, biblical) – is the apostle Paul, but at no point is his oral declaration of Roman citizenship questioned, no one ever checks on his status.43

Still, it is surely not irrelevant that the main term used to refer to the sixth, lowest, ‘class’ is ‘capite censi’, literally those ‘assessed by head’ (though ‘caput/head’ really signifies the legal concept of a person), not by property (on account of its insufficiency).44 This suggests that they did register: that the ideological significance of the census was as strong for them as anyone else, even if the practical ramifications were negligible; that they too were defined by their assessment by the state. There is also an interesting exception to the rule that citizenship is (p.183) generated outside the confines of the census, which adds a further dimension to the possible benefits of individual engagement with the process, even (or especially) at a distance from its upper echelons. This is in respect to one of the three standard modes of manumission, at least in the Republic, which are, as later jurists such as Gaius consistently list them, ‘by vindicta (a ritualized legal fiction), by census, or by will’ (Institutes 1.17, 35). If a slave rightfully owned by a Roman citizen makes a declaration for the census, with the agreement of his master, he becomes a Roman citizen; though it is a nice point, explains Cicero, whether his freedom commences at the moment his declaration is recorded, or at the moment of the lustrum, which completes – perhaps actualizes – the census (Orator 1.183). Either way, however, citizenship for these individuals is constituted through the census (censu); and it may be that a similar outcome is hoped for by another group of new citizens, or at least aspirants to that status, who also illustrate that registration, being on lists, could be of individual importance, as well as a key corporate enterprise.

Livy records a couple of occasions in the early second century BC when embassies from Latin colonies and Italian allies complained to the Roman senate that large numbers of their cives had migrated to Rome, and been registered (censa) there. In 187 BC, for example, a magistrate was charged with discovering these persons, and, if anyone, or their parents, was proved to have been registered in another Italian community reasonably recently (since the census of 202 BC), they would be sent back to their own, or their parental, community (Livy 39.3.4–6). Consequently, 12,000 were returned. The actual law in this area is vague, and debated, and it would be interesting to know what the content of the declarations these migrants made to the censors had been; there is no suggestion they were forsworn, that they had in any way lied under oath (or they would have been in real trouble).45 Still it seems clear that the incomers considered that registration would help establish, if not constitute, their change of citizenship – make them Roman; and, though the census turned out not to be constitutive, or, at least, that their transmutation was reversible, they were not completely wrong. The same action prior to 202 BC would have had a different result, and if they could not be found on another list, they stayed Roman.

As these events also show, registration of their own citizens was a reasonably common feature of Italian communities outside Rome at this time.46 This is hardly surprising, both in terms of the broader patterns of Mediterranean practice and the particularities of their relations with Rome. Most Italian states were obliged to provide troops for Rome, by the terms of the various uneven treaties imposed in the course of the Roman conquest of the peninsula, or, in the case of the Latin colonies Rome had herself founded to support her domination of Italy, by the terms (p.184) of their foundation. Whenever she needed armies, Rome ordered a levy of her own citizens and that the Latins and allies supply soldiery according to the formula togatorum, the means by which Rome determined the number each city or people was to produce. Much is unclear about the formula, but it seems that, having decided the overall figure required, this was then divided amongst the relevant states in rough proportion to their manpower, or at least Rome’s understanding of their young manpower, their number of iuniores.47 The Latins and allies then organized their own drafts, on a local basis, and making use of their own censuses, in the case of Latin colonies where institutions were modelled on those of the mother city, or looser registers, such as the cumulative lists of citizens kept by the Greek cities of southern Italy, such as Heraclea, before the Social War.48 The specific difficulties created by migration to Rome in this situation, implicit in the episode of 187 BC, are made explicit, rhetorically exaggerated even, in repeat complaints made to the Roman senate by Latin and Italian embassies just a decade later. In 177 BC they reported that 4,000 families had moved from amongst the Paeligni and Samnites (federated tribal states in central Italy) to the Latin colony of Fregellae, though no reduction had been made to their military levies (Livy 41.8.5–12), while resource problems were also again being caused by their citizens moving to, and registering in, Rome: so that, they alleged, with a flourish, ‘in just a few lustra, their now deserted towns and fields would not be able to furnish a single soldier’. The senate agreed measures to stop and even to attempt to reverse these processes (Livy 42.10.2).

It is interesting to note that, though these were, in some senses, Roman resources, there was no central effort to keep track of them. It was the Italian communities themselves who had to bring change, crisis even, to the attention of Rome, which otherwise assumed a steady state. Indeed, the response to this movement was attempted reversal, to stay with the formula. The point is further emphasized, with a twist, by the particular case of the twelve Latin colonies which, in 209 BC, after eight years of war with Hannibal, claimed total exhaustion and refused to provide Rome with any more men or money. Once Roman fortunes were restored, and the Carthaginians were on the retreat, in 204 BC, the colonies were punished with double levies, an annual tax, and a census to be conducted according to the formula issued by the censors in Rome, which would be sworn by the colonial censors, and brought to Rome (Livy 29.15.9–10). This the twelve colonies duly did, and the Roman censors received these local censuses directly after they had completed the lustrum; something which ‘had never happened before’, according to Livy, but was to ensure (p.185) that their military and financial resources were now a matter of permanent public record. Rome’s treatment of the twelve recusant colonies, the penal use of the census, imposing its terms and centralizing its report, with the shift in ownership of the relevant information that results, can be seen as foreshadowing a more systematic imperial practice from the time of Augustus onwards: the use of provincial censuses to assess the resources now available for exploitation by the Roman rulers.

The key event before that was the settlement following the Social War, and the grant of Roman citizenship to all Italian communities south of the Po. It was Julius Caesar, as dictator in 45 BC, who seems to have been the first to adopt a systematic approach to the new administrative landscape of Italy, but he was building on various aspects of previous practice, perhaps including various ad hoc recent developments. There were also some points of departure at least implicit in the structure which was now established. Basically, as the bronze inscription of the Table of Heraclea sets out, within sixty days of learning that a census was being conducted in Rome, the highest local magistrate had to conduct a census of the members of his community (who were now, as members, Roman citizens).49 He was to receive from all of them, under oath, their name, their father’s or patron’s name, their tribe, their age, and an account of their property, ‘according to the formula of the census which has been published at Rome by whoever is then about to conduct the census of the people’; and he was to ensure that all this was reported in the public records (tabulae publicae) of his municipality (lines 146–149). Following their compilation, these books were then to be sent to Rome, carried by officially selected envoys, who must present them to those conducting the census at Rome at least sixty days before the close of the census of the people. The magistrates at Rome, censors or otherwise, had five days to ensure that the contents of these books were honestly and accurately copied into their public records, and were stored in the same place as all the rest of the tabulae, ‘in which the census of the people is written out’ (line 156). The law also provided for those with multiple domiciles, and who had already registered at Rome, not to have to repeat the process locally.

There is no suggestion here, as there is in some accounts of the earlier censuses in the city of Rome itself, that fathers might declare their children, and husbands their wives. Declarations were to be received from all citizens, and, by the early imperial period, certainly, women seem to be registered in exactly the same way as men.50 This is, explicitly, an individualized, and apparently standardized, kind of registration. In addition, the system represented by the inscription from Heraclea (p.186) seems to have lost the moral dimension, the regimen morum,so crucial to the Roman census up till then. What is being transmitted to Rome, accurately transmitted, is information, not judgement; information that includes an assessment of property, according to Roman rules, but nothing more. One of the issues at stake in registering at Rome was, in some sense, inviting full censorial scrutiny, which would become increasingly limited to the real elite, while legislation of wider scope took on other aspects of the censors’ role. It is hard to think how it could, practically, have been otherwise, in this world of wide-spread citizenship, when Rome had to give up even the pretence of being a city-state; but the significance of the shifts involved should not be underestimated, shifts which were then deepened and solidified under Augustus.

These were not, however, moves towards a more utilitarian, statistical, census practice in relation to Roman citizens, at least as events turned out. What Julius Caesar’s plans were remain obscure, since his assassination in 44 BC brought his programme of administrative reform to an abrupt halt, and though his great-nephew, and adopted son and heir, Augustus, picked up some of the pieces after 31 BC, whether the way he put Roman citizen, provincial subject and imperial state together, his evolving fiscal, military and political settlement, bears much relationship to that previously envisioned is uncertain. Three key Augustan developments are most significant in this respect: the falling away of political participation except amongst the elite, as elections were re-structured (and eventually, almost immediately after Augustus’ death, passed to the senate: Tacitus, Annals 1.15); the ‘professionalization’ of the Roman army, so that military service became, by and large, a career choice not an obligation; and the placing of the burden of direct taxation on the provinces – tributum was now paid by those outside Italy.51 So, in his emphatic revival of the citizen census and lustrum, now more thorough in its registration of cives Romani in Italy and beyond, Augustus was not looking to create the fullest lists of those liable for military and monetary levies. That kind of resource management was more a matter for the emerging provincial censuses – such as that of Quirinus in Syria – which catalogued non-citizens, their wealth, property, and other features, by city or other administrative unit.

Not that citizen status was now entirely disassociated from the military and fiscal organization of the state: Roman citizens enrolled in the legions (or the elite praetorian guards), which had more advantageous terms of service than the non-citizen auxiliary units, and enjoyed some tax exemptions on Italian property, as well as broader legal privileges. But Augustus’ interests were focused (as was (p.187) traditional) on the upper echelons of the citizen census, on the centrally produced hierarchy it generated, and the overall magnitude of the citizen-body. The new princeps differentiated the top class further, and more strongly, than before, and utilized the basic structure of the census in building a joined-up social order right across the empire.52 The figures he proclaimed rose over the course of his principate: in Augustus’ third and final lustrum, conducted with his adopted son and heir, Tiberius, in the last year of his life, AD 14, the number of Roman citizens assessed was declared to be 4,937,000 (Res Gestae 8.4).

Thus, distinctions both between Roman citizens and others, and within these categories, were ideologically meaningful, and practically significant, for the maintenance and functioning of the empire, for its ruler, its ruling elite, and its wider populations. Taxation and other forms of resource extraction, and management, were part of this significance, but that was never going to be all that was at stake in these shifting encounters between people and state. Registrations of privilege as well as of exploitation would persist and interweave over the centuries; one of the markers of Roman statehood would continue to be the articulation of the two, the arbitration of different categories of belonging.


Despite the various differences in form and content, the registration systems of democratic Athens and Republican Rome share some key features. Neither is definitive of status. Identity is basically determined elsewhere, and is then meant to be recorded on deme registers and census rolls. For Rome this is a record of full members only, while for Athens, both the fullest and the most partial of participants in the state (taking slaves to be totally excluded) are listed. Both processes are public affairs. In Rome the professio takes centre stage – the individual declaration to authority – and the audience of (rough) peers, those who are there to make their professions, too, provide a generic form of validation by their presence: that they hear what is said, what is claimed. In Athens, on the other hand, the action is more collective: there is no individual declaration and no audience. Claims emerge from the social and political structure of the city: seventeen years after the ritual introduction of any newborn boys to their family and the larger social grouping of their father’s phatry (‘brotherhood’), the survivors will be considered in the deme assemblies. The deme debates and the deme decides, with some central oversight. The demesmen make themselves witnesses to citizenship, to the entry into full politeia, in a far more important way than going on a list does. While for the Romans not only personal testimony but also being listed count; they see strengths and (p.188) weaknesses in both as forms of proof. However, Athens and Rome converge again in their positive presumption that, if someone appears at the relevant deme meeting, if they present themselves at the census, that is part of behaving like a citizen, which needs to be disproved, rather than the reverse. Indeed, in these communities more broadly, appropriate behaviour does a lot of work in the formation and recognition of identity anyway.

The commonalities and divergences in their registrative projects and practices can in many ways be correlated to larger-scale similarities and differences between Athens and Rome in terms of state formation and political structure, and historical and social development. They can also, in the context of this volume, be located within much broader patterns of convergence and variety between polities and communities across time and space. This is a pattern, moreover, which clearly and importantly reaches back far beyond modernity in both East and West, and many places in between. Civil registration is a key feature of many ancient states, democratic, oligarchic, and monarchic; large and small; city-states and empires; not in exactly the same way, taking the same form, of course, but always about membership and privilege as well as subordination and exploitation.


Bibliography references:

Akrigg, B. 2011. ‘Demogaphy and Classical Athens.’ In Demography and the Graeco-Roman World: New Insights and Approaches, ed. C. Holleran and A. Pudsey. Cambridge: Cambridge University Press, 37–59.

Astin, A. E. 1988. ‘Regimen Morum.’ Journal for Roman Studies 78: 14–34.

Baronowski, D. W. 1984. ‘The Formula Togatorum.’ Historia 33: 248–252.

Blok, J. 2009. ‘Perikles’ Citizenship Law: A New Perspective.’ Historia 58: 141–170.

Broadhead, W. 2001. ‘Rome’s Migration Policy and the So-Called Ius Migrandi.’ Cahiers Glotz 12: 69–89.

Brunt, P. 1987. Italian Manpower: 225 BC–AD 14, rev. edn. Oxford: Oxford University Press.

—— 1990. ‘The Revenues of Rome.’ In Roman Imperial Themes. Oxford: Oxford University Press.

Clarysse, W. and D. Thompson. 2006. Counting the People in Hellenistic Egypt, 2 vols. Cambridge: Cambridge University Press.

Crawford, M. H. 1996. Roman Statutes, 2 vols. London: Institute of Classical Studies.

Crook, J. A. 1996. ‘Political History, 30 BC to AD 14.’ In The Cambridge Ancient History: The Augustan Empire, 43 BC–AD 69, 2nd edn, ed. A. K. Bowman, E. Champlin and A. Lintott. Cambridge: Cambridge University Press, 70–112.

Gardner, J. 1986. ‘Proofs of Status in the Roman World.’ Bulletin of the Institute of Classical Studies 33: 1–14.

—— 1993. Being a Roman Citizen. London: Routledge.

Grieve, L. 1983. ‘Tabulae Caeritum.’ In Studies in Latin Literature and Roman History III, ed. C. Deroux. Brussels: Éditions Latomus, 26–43. (p.189)

Hin, S. 2008. ‘Counting Romans.’ In People, Land, and Politics: Demographic Developments and the Transformation of Roman Italy, 300 BC–AD 14, ed. L. de Ligt and S. Northwood. Leiden: Brill, 187–238.

Kapparis, K. 2005. ‘Immigration and Citizenship Procedures in Athenian Law.’ Revue Internationale des Droits de’Antiquité 52: 71–113.

Keppie, L. 1996. ‘The Army and the Navy.’ In The Cambridge Ancient History: The Augustan Empire, 43 BC–AD 69, 2nd edn, ed. A. K. Bowman, E. Champlin and A. Lintott. Cambridge: Cambridge University Press, 371–395.

Launaro, A. 2011. Peasants and Slaves: The Rural Population of Roman Italy (200 BC to AD 100). Cambridge: Cambridge University Press.

Lintott, A. 1972. ‘Provocatio from the Struggle of the Orders to the Principate.’ Aufstieg und Niedergang der römischen Welt 1.2. Berlin: Walter de Gruyter, 226–267.

Meyer, E. A. 2004. Legitimacy and Law in the Roman World: Tabulae in Roman Belief and Practice. Cambridge: Cambridge University Press.

—— 2010. Metics and the Athenian Phialai-Inscriptions: A Study in Athenian Epigraphy and Law. Wiesbaden: Franz Steiner Verlag.

Moatti, C. 2009. ‘Le traitement des absents à Rome à l’époque républicaine et au début de l’empire: quelques considérations.’ In Le monde de l’itinérance en Méditerranée de l’antiquité à l’époque moderne: procédures de contrôle et d’identification, ed. C. Moatti, W. Kaiser and C. Pébarthe. Paris: Diffusion de Boccard, 321–349.

Mouritsen, M. 2001. Plebs and Politics in the Late Roman Republic. Cambridge: Cambridge University Press.

Nicolet, C. 1976. Tributum. Recherches sur la fiscalité directe sous la République Romaine. Bonn: Habelt.

—— 1980. The World of the Citizen in Republican Rome, trans. P. S. Falla. London: Batsford.

—— 1984. ‘Augustus, Government and the Propertied Class.’ In Caesar Augustus: Seven Aspects, ed. F. Millar and E. Segal. Oxford: Oxford University Press, 89–128.

Northwood, S. 2008. ‘Census and Tributum.’ In People, Land, and Politics: Demographic Developments and the Transformation of Roman Italy, 300 BC–AD 14, ed. L. de Ligt and S. Northwood. Leiden: Brill, 257–270.

Oliver, G. J. 2007. War, Food, and Politics in Early Hellenistic Athens. Oxford: Oxford University Press.

Osborne, R. 1997. ‘Law, the Democratic Citizen and the Representation of Women in Classical Athens.’ Past and Present 135: 3–33.

Patterson, C. 1986. ‘Hai Attikai: The Other Athenians.’ Helios 13: 49–67.

—— 2000. ‘The Hospitality of Athenian Justice: The Metic in Court.’ In Law and Social Status in Classical Athens, ed. V. Hunter and J. Edmondson. Oxford: Oxford University Press, 93–112.

Rathbone, D. W. 1993. ‘Egypt, Augustus and Roman Taxation.’ Cahiers Glotz 4: 81–112.

—— 1996. ‘The Imperial Finances.’ In The Cambridge Ancient History: The Augustan Empire, 43 BC–AD 69, 2nd edn, ed. A. K. Bowman, E. Champlin and A. Lintott. Cambridge: Cambridge University Press, 309–323.

Scafuro, A. C. 1994. ‘Witnessing and False Witnessing: Proving Citizenship and Kin Identity in Fourth-Century Athens.’ In Athenian Identity and Civic Ideology, ed. A. L. Boegehold and A. C. Scafuro. Baltimore, MD: Johns Hopkins University Press, 156–198. (p.190)

Scheidel, W. 2008. ‘Roman Population Size: The Logic of the Debate.’ In People, Land, and Politics: Demographic Developments and the Transformation of Roman Italy, 300 BC–AD 14, ed. L. de Ligt and S. Northwood. Leiden: Brill, 17–70.

Sherwin-White, A. N. 1973. The Roman Citizenship, 2nd edn. Oxford: Oxford University Press.

Suolahti, J. 1963. The Roman Censors: A Study on Social Structure. Helsinki Suomalainen Tiedakatemia.

Whitehead, D. 1977. The Ideology of the Athenian Metic. Cambridge: Cambridge Philological Society.

—— 1986. The Demes of Attica, 508/7–c.250 BC: A Political and Social Study. Princeton, NJ: Princeton University Press.

Zelnick-Abramovitz, R. 2005. Not Wholly Free: The Concept of Manumission and the Status of Manumitted Slaves in the Ancient Greek World. Leiden: Brill.


(1) Josephus, Jewish Antiquities 17.355, 18.2 and 26; Inscriptiones Latinae Selectae 2683. This, of course, is rather chronologically problematic for Luke; and, indeed, there are other difficulties with his account, too, over which much ink has been spilt.

(2) And probably others, though dated evidence is scarce: see Brunt (1990, 345–346), for a collection of the relevant material. Augustus’ successors continued this practice. For the difficult case of Egypt see (Rathbone 1993).

(3) The Romans, of course, extracted payment from their provinces before Augustus, under the Republic, usually in the form of reparations and through tax farming; Augustus regularized the system, but diversity persisted in various ways.

(4) Ptolemy I, a leading general of Alexander the Great (Alexander III of Macedon), established his rule of Egypt as Alexander’s vast empire was broken up into successor kingdoms following his death in 323 BC. The last of the dynasty was Cleopatra VII, defeated (with Marc Antony) by Augustus at Actium in 31 BC; Egypt was then annexed by Rome. On registration and enumeration in Ptolemaic Egypt, and their precedents, see the magisterial study of Clarysse and Thompson (2006).

(5) The precise relationship between the Ptolemaic ‘halikê’ and the salt after which it is named is unclear, for it basically functioned as a capitation tax: see Clarysse and Thompson (2006, vol. 2: 36–44).

(6) It is unclear to what extent foreign residents (that is, citizens of other cities who migrated to Athens) and freed slaves were placed in the same category as ‘metics’; the status and situation of these two, free(-ish) non-citizen resident groups certainly had much in common, but distinctions are also made, and have been recently emphasized in Zelnick-Abramovitz (2005, esp. 4–5, 99–107, 308–315). Even if manumission did not automatically make someone a metic, however, there does seem to have been a drift in that direction, and it should be remembered that the descendants of freed slaves in Athens inherited that status in some way – they did not become citizens unless an explicit grant was made – so a loose, encompassing grouping will have resulted, and it seems to have taken the ‘metic’ title.

(7) Whitehead (1986) is the classic study of the deme system.

(8) Athenian indifference to enumeration makes population estimates tricky, but most come in the20,000–35,000 range for adult male citizens, probably peaking in the fifth century BC, and declininginto the fourth. For a recent summary of the debate, with bibliography, see, for example, (Oliver 2007,78–87) and (Akrigg 2011).

(9) Aristotle, Ath. Pol. 42.1–2. This passage reflects the ambiguities of ‘citizenship’ in Greek states, which were usually, of course, not democratic. Belonging to a city was generally a status acquired through descent, and always brought with it a share in the state’s ‘hiera kai hosia’, cult and civic practices, for both men and women, but not necessarily political participation. It entailed rights of property and inheritance, as well as of involvement in the exchange between the human community and the gods according to ancestral tradition, but perhaps no more than that. Aristotle discusses these problems of definition at the beginning of Book 3 of the Politics (1274b–1278b).

(10) See (Blok 2009) for a recent and full discussion of the law.

(11) Whitehead (1986, 103).

(12) The reasons for Demophilus’ decree ordering this universal scrutiny are unclear; for discussion of the resulting law cases see Whitehead (1986, 104–109) and (Kapparis 2005).

(13) On the priority of living witnesses in these matters, see especially (Scafuro 1994).

(14) On female citizenship in general see (Patterson 1986) and (Osborne 1997).

(15) The precise role of the prostates, who had to be the former master for a freed slave in Athens, is unclear. Unlike in some other Greek cities, they did not actually have to act for the metic in court, but perhaps functioned as the witness of their status and identity, and vouched for their standing in some way. See (Whitehead 1977, 91) and Zelnick-Abramovitz (2005, 248–262).

(16) Thus, one later Greek lexicographer defines a metic as ‘one who pays the metoikion’ (Pollux 3.55); see also (Whitehead 1977, 75–77). Given that a skilled worker in Athens might earn around a drachma a day, the tax is not heavy, but the contrast with the (at least directly) un-taxed citizens is emphasized by the practice of state payment for a range of civic activities, from jury- to military-service. On freed slaves, whose military liabilities are also debatable, see Zelnick-Abramovitz (2005, 308–315).

(17) (Whitehead 1977, 77) suggests that there was a central register kept by the office of the poletai, orthe polemarch, the official with judicial responsibility for metics, but he admits this is speculative.

(18) The Macedonian hegemony of Greece was established by Philip II (Alexander the Great’s father) in338 BC, and brought Athenian democracy, at least as an autonomous entity, to an end (after Macedoniandomination comes Roman rule). Wealth restrictions on political participation followed, and Demetrius’population count (reported at Athenaeus 7.272b–c) is part of this reform process.

(19) Vulnerability much emphasized in recent scholarship: Patterson (2000) and Meyer (2010, esp. 28–47). Charges against a metic could be brought by anyone who wished to, and the incentive in such cases was that a successful prosecutor was entitled to a share of the proceeds: i.e. the sale of the metic and their property.

(20) Demosthenes 25.5.

(21) That tax receipts, rather than registers, might be available to be consulted is, however, suggested in another law case: Demosthenes 57.33–4.

(22) The classic study of the Roman citizenship and its expansion is Sherwin-White ([1939] 1973); see also Gardner (1993).

(23) Dionysius of Halicarnassus, Roman Antiquities 5.20.

(24) Livy, Periochae 20 and 98. A table of recorded figures for all Republican and Augustan lustra canbe found in Brunt ([1971]1987), 13; the discussion about the reliability of the figures is on pp. 26–43.

(25) See, for example, (Scheidel 2008), (Hin 2008), and (Launaro 2011, 25–50), for recent discussion.

(26) They went into the annual records kept by the pontifices (one of the priestly colleges in Rome) – the annales maximi – and from there, and through each other, into a range of annalistic and historicalwritings.

(27) Livy 1.42.4–5.

(28) Livy 1.44.2.

(29) For discussion of the Republican census as process, see (Suolahti 1963, 32–47); Brunt ([1971] 1987, 15–25); Nicolet (1980, 48–88); and (Northwood 2008, 257–265).

(30) Only Dionysius mentions declaring wives and children: Roman Antiquities 4.15.6 and 5.73.3 (including children’s ages); while Cicero makes a rather different kind of reference to the censors’ interest in assessing offspring (and ages): Laws 3.7. Questions about marital status, cast as standard, appear at Aulus Gellius 4.20.1 and 4.3.2 (also 17.21.44).

(31) Censorial control over valuation: e.g. Livy 39.44.2–3 (censorship of Cato); tribal movement of freedmen: e.g. Livy 45.15.1–6.

(32) There was a punitive grouping – the aerarii – in which persons could be placed. Its exact nature is unclear, but being listed as such certainly entailed ignominy and perhaps financial penalty, too; the votes of those classified as aerarii were cast in a special (non-)century, and so barely counted: the best discussion is (Grieve 1983). On the various causes of punishment see (Astin 1988).

(33) How any lists of individuals, rather than their classification into the relevant groups, played a role in this process is less clear, at least once Rome reached a significant size in the Republican period. There are reports for early Rome of a literal ‘roll-call’ (e.g. Livy 3.10.1), but by the time of the Hannibalic War it seems that going through the lists of iuniores (those aged between 17 and 45 and liable for regular military call-up) was not usual practice, since the censors of 214 BC do it only because they have timeon their hands and some particular concerns (Livy 24.18.7). On the dilectus generally, see Brunt ([1971]1987, 625–634).

(34) On tributum see (Nicolet 1976).

(35) This time writing his history of Rome in Greek, that is Dionysius of Halicarnassus (4.15.6)

(36) This is the position set out in the Oscan law partially preserved on a fragmentary bronze tablet fromBantia in southern Italy, and perhaps dating to the early first century BC (Crawford 1996). The textbriefly describes the local census (IV.9–10), but it is likely to be as indebted to Roman practice here aselsewhere, perhaps even adapting procedures from the nearby Latin colony of Venusia. This wouldcertainly be a practical approach, reserving the possibility of prosecution and punishment, if necessary:see (Moatti 2009).

(37) Cicero, Verrines 2.1.104–107.

(38) The stark contrast drawn by (Meyer 2004, esp. 12–43) between an Athenian legal system based entirely on personal testimony, and a Roman system where the document was king, is, therefore, much exaggerated; though there is more of a place for documents in Rome, and their role was to become greater under the empire. For a more nuanced view see (Gardner 1986).

(39) In the absence of the prosecution speech, of course, whether Archias’ absence from the census hadbeen explicitly raised, and in what way, is unknown; so Cicero might simply have been covering thebases rather than responding to a specific accusation. The result of the case is also unknown, thoughthe indications are, since Archias stays in Rome, that he (and Cicero) prevailed.

(41) See (Mouritsen 2001, 113).

(42) Brunt ([1971] 1987, 34); and on provocatio in general, see (Lintott 1972).

(43) Acts 22: 25–29.

(44) The other term is ‘proletarii’, apparently on account of the offspring (‘proles’) they contributed tothe state: Cicero, Republic 2.40; cf. Aulus Gellius 16.10.

(45) The best discussion of the law on migration to Rome is (Broadhead 2001).

(46) A prevalence already indicated by the law from Bantia, a small city with no particular political significance in its region.

(48) Cicero reports (Pro Archia 6 and 8) that Archias ‘was written into’ (ascriptus) the citizen-lists of various Italian cities, most importantly Heraclea, on his way to becoming a Roman citizen. This kind of register is rather different from the Roman census, however, lacking its more communal cyclical and ritual aspect, and the key feature of assessment.

(49) The Table of Heraclea (named after its place of discovery) contains a range of regulatory provisions from the last years of the Republic: see (Crawford 1996) for text, translation and discussion.

(50) Unfortunately this evidence is preserved in a list of the long-lived compiled by Phlegon of Tralles (Fragments of the Greek Historians 235 Fr. 37), so whether children were also included remains obscure.

(51) This is, of course, to simplify more complex patterns: citizen conscription was not entirely over, for instance; Augustus instituted a (rather restricted) inheritance tax on Roman citizens in Italy; and the popular element never entirely disappeared from Roman politics, transferring to the more informal setting of the circus and arena. On Augustus’ military, fiscal and political moves, see (Crook 1996), (Keppie 1996), and Rathbone (1996).

(52) On Augustus and the census, see Nicolet (1984).