Navigation – Plan du site
Codifications et réformes dans l'Empire tardif et les royaumes barbares
II - Perspectives diachroniques
La loi, instrument de réforme

Reform and Legislation in the Roman Empire

Boudewijn Sirks


Reformare means bringing something into a new or previous form, or change its nature. Here this is restricted to bringing about changes in social behaviour (social engineering). Social structures in Roman Antiquity must have been firm and consequently social rules too, but they were nevertheless subject to some change over the centuries. Further, legislation was, in Roman Antiquity, generally not used to introduce innovations. In the Republic there are therefore very few cases of such reforms, and these are all cases of the law as a last resort to maintain vanishing social rules by enforcing them. In the Empire the situation was not so much different. Yet, there are, in this period, more cases to be found in which a legal measure had a certain social impact and where a social change generated a legal response in order to counteract the undesired consequences. Though not spectacular, compared to the general stability of the social structures they are not negligeable.

Haut de page

Notes de la rédaction

Les textes réunis dans ce dossier par Olivier Huck constituent les actes de la table ronde organisée à l’École française de Rome les 30 juin et 1er juillet 2009, sous l’intitulé éponyme Codifications et réformes dans l’Empire tardif et les royaumes barbares.

Texte intégral

  • 1 OLD, s.v. reformo : to alter in appearance, form ; to alter in nature ; to restore to a previous fo (...)

1The theme of this conference, legislation and reform of the city and of the state, seems to me calling for a preliminary question. Can we speak of reform by legislation as we do in our times? What do we understand by this? First reform: it has a wide meaning. Fundamentally the word refers to bringing something into a new or previous form, or changes its nature,1 and what is brought into a new or previous form or changed can be anything. A person can be reformed and we understand by this that he changes his ideas, views, belief or whatever: he becomes a so-called « new » person. In criminal law one of the objects of imprisonment is indeed to change and if possible to improve the criminal. In the 16th century we see in the Dutch Republic, in London and some North-German cities houses of correction (« verbeterhuizen ») founded and a current idea was indeed to bring vagabonds to a better life, namely one of industry and diligence. Or it happens in religion, a process better known as conversion (« metanoia ») or repentance, depending on the circumstances. The effect, or recompense if one likes, is here the outlook to a better life in this or the next world.

  • 2 Such as nowadays : to bring about a change in behaviour by fines (e.g., fining throwing used chewin (...)
  • 3 See Mantovani 2012.

2Restricting ourselves to law, was there ever a general idea that legislation could be used to reform, that is, as a means of orientating a society towards new or previous goals or values, or of reinforcing the central administration, or as a means of social engineering or of a policy ?2 That depends first of all on the idea a society has of legislation. As regards Rome in Antiquity, legislation played a great part, both in the Republic as in the Principate, in law. The leges rogatae were statutes in the modern sense of the word, edicts were also in the modern sense legislation, be it as imposed rule, but here the validity originally did not outlast the tenure of their author. Notwithstanding the multitudo legum3, basically, the law was customary law. It was comprehensively rendered in the Twelve Tables, supplemented and refined by leges, plebiscites and edicts, and these were given effect in a limited number of actions in the Praetorian Edict. To this came the authoritative opinions of the iurisperiti, the senatusconsults and the decisions and interpretations of the Roman emperors. The Roman jurists listed, described and systematised this customary law and their renderings, encapsulated in their writings, became the unofficial (and later official) recording of the law (the ius vetus). This process of accumulation implies that we should not expect robust measures. The legislation of Sulla (138-78) could be considered an attempt to reform the state in the years 82-79 ; some of these were later repealed. However, he was an exceptional phenomenon and his measures were not aimed at changing social mores.

  • 4 The aedilician aediles imposed their edict on market sales. But in view of Varro’s formula’s for th (...)

3After the first century AD the position of the emperor led to so called ius novum, law originating in constitutions of the emperor only. Their unique position as supreme judge, de facto sole lawgiver and head of the state administration would have made possible for them to introduce profound changes by way of legislation, but they did not do so, notwithstanding that Constantine is called by a later source a pertubator rerum. The ius novum was rather a supplement to the ius vetus. The only notable exception are in later times the laws against heretics and religions competing with Christianity. Perhaps we may find with Justinian some bolder innovations, but that would be all. What we also see happen in the Later Empire is, that in case a law suit gave rise to a general rule or an adaptation of a general rule, the emperor issued a constitution (mentioned in CJ 1.14.2-3 ; an example is Nov. Sev. 2). In the west the legal situation returned to a more customary level with the disappearance of the western emperor and the introduction of barbarian customary law, which was duly recorded under the Visigoths, the Longobards and Charlemagne. In the High Middle Ages legislation still was a derivation of jurisdiction. The supreme judge would convert a judgment into a general rule if he considered it fit to be so. Or others would distinguish in a judgment a general rule. Incidentally princes were determined to set aside their feudal barons and establish a power base of their own and here we see something of legislation as an instrument (such as the Constitutions of Melfi), but still the installation of a royal central supreme court was up till the 17th century the preferred and best means to centralise a state. For a new conception to break through, viz. that legislation can be an instrument of change and consequently of reform, we have to wait till 1748. In that year Montesquieu published his De l’esprit des Loix, in which he proposed that laws are dependent of time, geography, climate etc. Reversely, it presented a basis to use positive law as an instrument to make law, adapted to what was conceived as special to a region or nation. After Montesquieu the wish for national codifications arose, and in the French Revolution the idea that by laws a social reform was possible was executed. Law could now be instrumentalised. Ever since governments have returned to this idea, and it is nowadays a current view that to a certain extent social behaviour can be manipulated by law. But here we are on a theoretical level. On a practical level, such as in towns, bylaws were already before Montesquieu used to organise public life. There are many urban « policy statutes » or « Polizey-Verordnungen », already in the Middle Ages, and well known are market regulations which certainly did not all just codify existing unwritten rules. And in that sense the Edict of the aedilician edicts, ruling the sale of slaves and cattle, might also already have been partly regulating and socially engineering.4

4After having discussed the preliminary question on legislation, the question now is what to understand by reform. Reforms can deal with many aspects of a society but I want to explore whether we can see it used for social change and social engineering. We cannot expect stunning social operators like nowadays statutes on emancipation of women or minorities, or a prohibition of smoking in public places, or prescription to wear safety belts or helmets when motoring. But we do see some « social engineering ».

  • 5 Berger 1953, s.v. sumptus. Such statutes were the leges Aemilia, Fannia, Oppia, Orchia and Iulia Su (...)

5Some statutes, particularly the sumptuary laws,5 were directed at a social change, but only in the sense that they were meant to stop social change. One may call these cases of social engineering, no more than legal means to enforce existing moral precepts of a small elite. Upon whom ? Sometimes upon the same elite, but in that case on the part of it which evidently no longer followed these precepts. Thus we see conservatives trying to stop others from change, or men trying to restrain women in new situations. E.g., the lex Oppia of 215 BC introduced for women restrictions on jewelry and forbade multicoloured dresses, but was abolished twenty years later by the lex Valeria Fundania : evidently a case where change could not be turned back. The Senatusconsultum de Bacchanalibus of 186 BC, prohibiting the mystic festival of the Bacchanalia, is a more famous and wider ranging example of this application. These laws are used where the ordinary means to uphold the morality of old, such as paternal authority and notae censoriae, no longer have effect. Law is here the last resort for moral enforcement.

  • 6 Buckland 1996, p. 292-293, 319 ; Kaser 1971, p. 318-321, 724.
  • 7 Capogrossi C. 1981, p. 3-12 ; Capogrossi C. 1988, p. 275ff, 267ff.
  • 8 Franciosi 1975, p. 57-68.

6The most famous case in Antiquity of a statute said to turn around a social change which already had taken place is Augustus’ marriage law, supplemented with the lex Papia Poppaea.6 It is often assumed that it meant to rekindle the wish for families and offspring by limiting the capacity to enjoy by testament. Yet, at a closer look, it is not such a simple case. Within six grades of kinship there were no restrictions in inheriting at all. What the law clearly really wanted to restrict was the bequeathing beyond this grade, that is, making assets going out of the control of the near kinship. It was, therefore, meant to strengthen family fortunes and this was sensible, because senators and decurions needed capital to function in public life and new members of these groups were predominantly chosen from amongst their descendants. Thus it intended to put a stop to a social change, namely to bequeath to strangers (extranei) without a family future – apparently an upcoming trend. They could be endowed, but only if they themselves were a nucleus of a future family. What the law, in the end, envisaged was to keep fortune and family together : it was a means to enforce a specific, very traditional form of family strategy. We see legislation used to counteract social changes, hence used reactively, and again to enforce vanishing social rules. Besides, the limit of the sixth grade may also point to another social phenomenon : that the greater family contracted and people felt less connected with far removed relatives. Capogrossi Colognesi has posited a similar shift for the fourth century BC from gens to familia,7 and here we may be confronted with a further shift within the familia. Likewise Franciosi concludes that Augustus’ restriction for senators and their children to marry a freed person reflected the original non-existence of conubium between freeborn and freed persons.8 This again is an attempt to restrict a social change. Augustus’ allowance for other ingenui to marry freed persons will have been rather the condoning of an existing practice than making the hitherto impossible possible. The lex is in any case never read in this way.

  • 9 Buckland 1966, p. 283, 284-285 ; Kaser 1971, p. 106-108, 678-679.
  • 10 See, e.g., Sirks 1994, p. 273-296.
  • 11 Buckland 1966 (3rd. ed.), p. 328 ; Kaser 1971, p. 709-712. It was done by way of the querela inoffi (...)

7Other legislation might mean to correct abuses or potential dangers. A case of this is the legitimate portion for children of the testator. Originally the Roman law of succession was intestate. The testamentum calatis comitiis was rather a suspended adrogatio, to become effective upon the death of the testator. Then, in the third century BC, the testamentum per aes et libram, the mancipation testament, was introduced, presumably to make bequests and a division among the intestate heirs possible. Only if there were no heredes sui or agnati it could be used to appoint an extraneus as heir.9 Till so far the basis of intestate succession, preferably by descendants (likely prompted by the wish to have the sacra familiaria continued and be assured of one’s own cult as ancestor).10 But its application was extended : it was also used to disinherit children and the wife in manu and institute extranei as heir. Almost at once, so to speak, the praetors reacted with the introduction in the first century BC of the legitimate portion for the intestate heirs of the de cuius.11 It would be interesting to connect this with the previous example and explore a change in th family structure, but the outcome in both cases would not change : the change occurred and the legislator tried to correct the outcome of this in a conservative sense.

  • 12 See Buckland 1966, p. 372-373.
  • 13 This senatusconsult was made for mothers not in manu ; otherwise they would have had a claim as agn (...)

8Again, legislation often followed a change in public opinion. E.g., the Tertullian (under Hadrian) and Orphitian (178 AD) senatusconsults invested the cognatic relation between mothers and children with some rights. The first gave a mother under conditions a claim to her children’s inheritance ; the second children a claim to their mother’s inheritance.12 They testify to a loosening of the agnatic structure of the family and to a strengthening of the position and independence within marriage, and perhaps even in the case of the Tertullian senatusconsult to an increase of marriages sine manu,13 but still they rather follow than precede such a change. Of course they would have facilitated further change, e.g. in those social strata where marriages cum manu were still the rule. But to assume a feat of social engineering here seems to me unwarranted.

  • 14 See Sirks 1991, p. 41-42, 61-67.

9A case in which legislation was meant to change a habit and to introduce something new were Claudius’ measures in 51 AD to have ships built and to stimulate sailing during the winter season. A shortage of grain had brought him in the preceding winter into political trouble. In order to increase the import of grain he offered legal and financial advantages to those who had ships built and used for the importation on behalf of the Roman market. Further, he took upon him the losses due to shipwrecking in the winter season, naturally when incurred with grain transports for the Roman market. Usually one did not sail in the winter season because of the storms.14 The latter measure did not have the effect wished for ; the first seems to have been effective. But it was the latter measure which really would have led to a behavioural change.

10What about the Later Empire ? To begin : did the imperial legislation, as found in the Theodosian and Justinian’s Code, have much contact with reality, or not ? It is likely it did. As the Lex Citandi of 426 AD says (in CJ 1.14.2-3), much legislation was enacted as a reply to questions from lower functionaries, or as a reaction on reports from the lower echelons of the administration, or in line with a judgment – where the rescript and appeal system ensured that even individuals on the lowest level could get through to the emperor. And a mere look at the Codes show that generally these enactments are reactions to situations in real life. Perhaps some emperors idled their time away in the pursuit of pleasure, perhaps some prefects and officials too, but overall the imperial administration had a good view on what happened in society, including the frauds by which people tried to hide their misdeeds.

11One might think that though this may be true, its interventions were ineffective. Here were have to be careful because we lack the means to check this and it depends on the way we interpret the Theodosian Code as source. But there is a factor which I want to mention here. Effectiveness depends on the measure that one needs to be effective to achieve a goal and on what one has as goal. If an administration sets out to achieve its goals by way of self-perpetuating social and economic machineries, it has, once these machineries are well set up and in motion, little to do. Some adjusting may be necessary but that would be all. I would submit that much of the administration in the Empire, particularly the Late Empire (which in this respect starts for me under the Severi) was done, from the point of view of the emperors, by such machinery and that the legislation we see enacted on this is fringe legislation. For some three centuries all went reasonably well : city administration and government, public distributions in the four big cities of the empire and its supply, taxation and its levying. I do not say that it was as good at the end of this period as it was in the beginning, but 300 years is a long time, much longer than many modern administrations have existed. And are we nowadays all happy with our municipal and national governments and administrations ? How effective is our legislation ? Roman imperial government experienced indeed much turmoil in the third century, but administration went steady on and the accomplishment of the Roman emperors to provide for a reasonably stable administration over such an enormous area is really admirable.

12Also reactive, but nevertheless an attempt to change behaviour, were, in the third century, the persecutions of the Christians. Emperors tried to force Christians into acceptable social behaviour, viz. into making an sacrifice to the emperor, and had them punished them if they did not. We know that they sometimes succeeded, sometimes not. But the Christians formed a small minority in those days and their behaviour will been considered asocial by others, and these persecutions may have attracted more attention due to the later dominant position of Christianity than they might have done otherwise.

  • 15 See Millar 1992, p. 3-12, 203-272, 636-652.
  • 16 Schmidt-Hofner 2008, p. 337-344, summarising his research.

13But were the emperors always reacting in a casuistic way to changes, as Millar has proposed ?15 Are there no cases of reform or attempts to this ? One way to look at this is to consider the frequency of constitutions. Most sources for the fourth century are present in the Theodosian Code. The constitutions are widely spread over the years, with sizeable peaks under the reign of Valentinian and Valens. Does this increase mean that these two emperors used legislation to change things ? Or did they merely react more often ? The latter would imply a higher frequency of action under them, which is not impossible (an emperor who is known to facilitate litigation and to judge fairly, will likely attract more claims). Apart from the impossibility to solve the question, whether this is statistically an acceptable representation anyway – it nevertheless remains an impressive number and it is indeed plausible that these emperors did issue relatively more constitutions than those before or after them – does this necessarily mean that change was the purport of this ? Schmidt-Hofner has indeed argued that, although Millar’s thesis of casuistic reactive imperial legislation remains valid for the overall picture, we see in the years 364-375 that great reforms took place. They required careful planning in advance and had a great impact : the reforms of the coinage, of the quotation system in taxation for the towns, of the supply for the limes, and other reforms, these had far going consequences for all involved and led to structural changes. Next to that the two brothers, Valentinian and Valens, were also interested in affirming their position as homines novi against an established order. Legislation could also serve a propagandistic goal : to underline that there was an emperor and that he was very and everywhere present.16 This is certainly an aspect we have to keep in mind. Legislation partly served the same propagandistic goal as the mint and the free distributions. But this should not lead away from Schmidt-Hofner’s argument, that in those years structural changes in the Roman state were undertaken and formulated by statutes, and that, if we follow his characterisation, la cité était réformée.

  • 17 This is not to suggest that Schmidt-Hofner might be wrong, but for the present argument the questio (...)
  • 18 Here we see a reform by the introduction of the new rules which established the Tetrarchy and as su (...)
  • 19 Beaucamp 2007, p. 275-285.
  • 20 CJ 5.17.8, of 449.
  • 21 Kaser 1975, p. 177.
  • 22 This does not mean that the interest in sermons, or religion, would not have been genuine. But thes (...)
  • 23 See on the changes in Egypt, our best documented case : Bagnall 1993, passim and part. p. 195, 198- (...)

14Yet, restricting myself to looking for a connection between legislation and social change, it is not certain that the reforms, or in any case legislative measures (taking as legislation every general rule issued) as signalled by Schmidt-Hofner, had as consequence a social change in the sense of a change in human behaviour in relation to other humans. Assuming for a moment that these were indeed reforms,17 a reform in taxation or minting may still leave social structures intact and unchanged. As said above, the position of the emperor after the first century AD would have made it possible for him, as sole legislator and supreme judge, to enforce changes by legislation. After the constitutional reforms of Diocletian his position was even stronger.18 I further distinguish between the short and the long run. The latter leads to the question, what impact the general grant of Roman citizenship in 212 AD had. It appears that by the end of the Late Empire people generally applied the Roman legal norms, but not always entirely. And regarding the restrictions set to divorce by Constantine and Constantius II in resp. 331 and 342, these were not applied at all.19 It is unknown what influenced these emperors to introduce these, but a Christian influence is not likely, as may, on the contrary, have been the case with the measures of Theodosius II in 449,20 or with those of Justinian, who may have been more successful.21 Another case in this category is the presumption that Christianity had an impact on legislation in order to realise Christian values in society. Here we see indeed some interventions to restrain or prohibit theatre (CTh 15.5), obscene plays (restriction on the maiuma : CTh 15.6) and gladiatorial games (CTh 15.12), which, we can be sure, disappointed many. These are actually short run successes : it is hard to believe that the desire for these kinds of entertainment was now for the pagans satisfied by the alternative of a weekly sermon.22 There is no sign that in Late Antiquity general social mores became Christian or improved as consequence of this legislation, notwithstanding that the role of the Church became more prominent and more and more people became Christian. A new social phenomenon was monasticism : living in seclusion and chastity in that fashion was not practised before.23 But this was not induced by law.

  • 24 The literature on this is extensive. See for this, as to the connection with the Council of Ephesus (...)

15A notable exception forms the late antique legislation which tried to stamp out religious competition. The Roman emperors since Theodosius I tried by stick and carrot to turn dissidents and others into the direction of the orthodox Christian faith and the Church. But where previously emperors up till Constantine tried to bring Christians to sacrifice to the emperor by force and mere application of the law on high treason, their methods were new. They issued constitutions restricting the legal position of heretics. Comparable are the laws on the Jews, who were not forced to convert, nor stimulated to do so, but rather were restrained from proselytising. Arians and Judaism seem to have been forceful competitors to the orthodoxy in the fourth century. But apparently most feared were Manichaeans and the many Christian heresies and against these rather severe measures were enacted, meant to coax the dissenters into the state church. Book 16 of the Theodosian Code contains many of such laws.24 Of course their conventicles were forbidden : thus CTh 16.5.34 of 398, e.g., orders to expel from the cities Eunomian and Montanist clerics and if they try to set up congregations in the country side, to deport them. But other methods were not eschewed. CTh 16.5.40, of February 407, puts Frygae and Priscillianists on one line with Manichaeans and orders that their assets are to be transferred to close relatives, unless these are of the same abominable belief ; and they lose the faculty to make private law contracts. Likewise they cannot make testaments, and their heirs ab intestato cannot inherit if they are not orthodox. On the contrary, CTh 16.5.41, of Nov. 407, offers heretics, Donatists and Manichaeans full pardon if they repent and revoke their errors. Manichaeism was particularly targeted. CTh 16.5.18 of 389 orders to drive away Manichaeans from Rome and confiscate their property : they cannot leave it by testament or intestately (Nihil ad summum his sit commune cum mundo, closes the text, conveniently summing this up). CTh 16.5.17 of 389, deprived likewise the Eunomians from the active and passive faculty to testate, but CTh 16.5.23 of 394 restored this to them, probably only in the east. Remarkably enough, paganism did not attract so much corrective attention. Also Arianism received a benign treatment in this respect, probably on account of its strong presence amongst the Goths.

  • 25 E.g., Donatism had affected entire villages in Africa. See on Donatism and its repression H. Chadwi (...)
  • 26 The first council where this shows is the Council of Sardica of 343, which established rules to enf (...)

16Here we see legislation used not merely to bring about an individual change but also, due to the extent of the phenomenon, as an instrument to bring about a considerable change in society and as an instrument of social engineering : succession was the major way of transferring assets. Blocking this would deprive children of heretics of their livelihood. It could finish communities in one generation.25 Was there a reason for this aggressive approach, so outside of the usual ? Were the emperors inspired here by the ecclesiastical synods and councils where since the faith was put into definitions and the creed formulated, to be adhered to by all true believers ? There these canons were indeed used since 343 to direct the mind and behaviour of the faithful and not so faithful and applied, if necessary, by way of condemnation and excommunication.26 Although previously emperors had tried to force Christians into acceptable social behaviour, viz. to make a sacrifice to the emperor, and punished them if they did not, they never had formulated such a body of special rules on this : they merely applied the law on treason. If this is the case, these constitutions of Book 16 form a special body of law within the entire body of law. Whether all this legislation had effect is not certain. One might say that the absence of many sects in the sixth century (particularly Manichaeism was stamped out) testify to their efficacy and this is certainly a good argument. But other approaches – conciliation, or, on the other hand, verbal attacks by Church Fathers, physical attacks by orthodox Christians – may have had effect as well and perhaps even better (although it could be said to have been endorsed to some extend by the imperial legislation against heretics) ; and other, new heresies may have drawn on old ones.

  • 27 CAH 1998, p. 371-388.
  • 28 Sirks 1996, p. 102.
  • 29 Laniado 2002, p. 134-149.

17The situation of the towns in the fourth century has been well described by Ward-Perkins.27 He concludes that the position of the town councillors was hollowed out by higher fiscal burdens and the shift of wealth to imperial administrators. Further, the importance of local administration was reduced by a more intensive imperial bureaucracy. This tallies very well with the substitution of the curia under Anastasius in some cases by the three or five largest landowners and the bishop, who often will have been a big landowner himself in his capacity as manager of his metropolitan church.28 Yet it does not necessarily imply social change. We do not know what role those large landowners already played in the councils or in the towns. It may have been merely a shift from the council to the category of notable landowners to another group. Laniado has indeed underlined the familial continuity here, next to some homines novi, migrated notables and imperial functionaries.29

18However, the introduction of the compulsion regarding the town administrations and the transportation for the public distributions, under the Severi, led to an unwanted social effect, in its turn combated by new legislation directed at social engineering. The compulsion implied that new councillors would be summoned from an eligible group of citizens : the legitimate sons of councillors, who had to dispose of sufficient capital to carry the expenses connected with the councillorship and the various duties deriving from it. Their status was high enough for the dignity of the council. The new system, meant to work once set into motion, as an independent machinery of local administration must have worked well. Why ? Because we see in the fourth century several means attested, by which councillors tried to escape the system.

  • 30 See Sirks 1991, p. 171-172 ; cases in which the naviculariate already had become an obligation pres (...)

19The earliest known and rather drastic means is the shift in marriage policy, attested in CTh 12.1.6 of 319. According to this town councillors preferred to live with slave girls instead of marrying free women of their own status, and consequently preferred to have their children born as slaves. Were these later on freed and thus of a lower social status and so unfit for the council ? The text suggests that they indeed tried to keep by this their descendants out of the councils, seeking protection of powerful people : it might have been the greater landowners or the imperial functionaries in the regional capitals. The fact that legislation was considered necessary means that the system worked so well that decurions did not see any other way to maintain their families and save their assets, and that it occurred enough to make legislation sensible. But was this legislation in its turn effective ? Did the harsh sanctions (forced labour in the mines for the woman, banishment for the man, confiscation of his property) have effect and crush out this new habit ? It would seem. Then, later on, new devises are mentioned, which retain the social standing. It suggests that the councillors had been looking for other ways of escape, which could suggest that the emperor had been successful. The new way was putting your land and consequently yourself by way of a feigned sale out of the reach of public burdens and enjoy the privileges of the coloni of the imperial res privata (CTh 12.1.33, 342). Presumably the sale was officially for a low price, diminishing the decurion’s fortune below the level required for the decurionate, while the paid true price was concealed. The lower status of a colonus (see below ; although coloni on the imperial estates ranked probably higher that those elsewhere) was, evidently, no objection. Still later on, in CTh 12.1.149 of 395, we see decurions in Africa taking the opportunity of an inheritance from a navicularius of the African corpus to shift their assets into this category of public obligations, by extending the exemption of the navicularii from the decurionate to their own patrimonial assets. By this they profited from a lesser overall burden. The assets from the navicularius they had to invest into a ship and let it make transports of grain to Rome under the general management of the corpus of Africa. Socially this move into the naviculariate was within bounds for them since these navicularii were of the same upper social stratum as they. The emperor reacted by separating in such a case the assets. The part from the naviculariate remained reserved for this duty, the other part for the decurionate. In this case the imperial policy took away the advantages, but its method was not so novel.30 The councillors remained now within their social stratum.

  • 31 See Sirks 2008, p. 124-128 on the interpretation of origo which made this tie possible.
  • 32 Condicionalis in CTh 8.1, 12.7.3, 14.9.2 ; the gynaecearii in CTh 10.20, the fabricenses in CTh 10. (...)

20The agricultural economy was of great importance for the Roman government as a source of fiscal revenues. Hence the abandonment of land, the so called agri deserti, caused great worry. One way to remedy was to sanction the colonate so that coloni were tied by public law to their assigned plot.31 Similarly those artisans working in the statal industries as the silk weaveries, the mint and the arms industries, and those in the murex fisheries, were tied to their occupation by the forced membership of a collegium ; as were those workers as city labourers, the collegiati. They, their descendants and sometimes in-laws were tied to their condicio, hence the designation of condicionales.32 This, however, led to a socially inferior, almost servile status, which was almost not remedied by the government. The so called colonate was in my opinion on the contrary an individual affair which, however, as with the condicionales, led to socially very low and subjected position from which many coloni tried to escape. The introduction of a general 30 year prescription in the east in 424 and in the west in June 449 (N. Val. 27) was an esthetically fine measure. But it had unwanted effects. Coloni originarii in the west now claimed freedom from their subjected status after 30 years. Valentinian quickly reacted. Already in January 451, one-and-a-half year later, he introduced also an acquisitive prescription for coloni (N. Val. 31.1). If they could claim after 30 years freedom from their former master, then somebody with whom the colonus had worked during the largest part of these years could claim him as his new colonus. Agriculture had priority. One-and-a-half year is rather quick for those days. In the east the drive for freedom was differently and more positively taken up. Anastasius introduced a change in their status after a service of 30 years : the colonus remained tied to the land but regained his personal freedom, which was considered a social advantage ; apparently he hoped it would encourage them to stay (CJ 11.48.19). We see that in the first case legislation led to an unwanted social effect, which was counteracted, whereas in the second case it was directed at a social effect of which benefits were expected.

21The abolition of the SC Claudianum in 531-534 (CJ 7.24.1 pr. ; 1 = CJ 11.48.24) could have an effect, clearly expected but certainly unwanted. Coloni adscripticii were no longer subjected to its analogue application and went out to marry free women. Contrary to before, children out of such a union were no longer subjected to the colonate but free from it. Justinian gave therefore in the same statute the master of the colonus the authority to reclaim the colonus and have him castigated (CJ It implies that the couple was separated forcibly. Here an unwanted social effect could be smothered in the crib.

  • 33 L.R. Burg. 37.2 : Quod si pares fuerint honestate persone, consensus perficit nuptias ; sic tamen, (...)

22Legislation could therefore lead to social effects and some measures may be called attempts to social engineering. But sometimes social customs were so deeply inveterated, that they never got a legal formulation. There is much law on the dowry, but the textbooks of Roman law state that a marriage was concluded merely by consent, unless accompanied for the confarreatio or coemptio by a formal act. Yet these merely ensured the direct establishment of the marital power over the woman. In reality the dowry was far more important : we may say that without dowry there was no marriage. However, that we never see formulated in any of the Roman legal sources. It is only in the lex Romana Burgundionum, drawn up in the early sixth century, that it is said in section 37.2 that consent is necessary for marriage, but that a marriage is not valid unless a dowry has been handed over or stipulated.33

Haut de page


Bagnall 1993 = R. Bagnall, Egypt in Late Antiquity, Princeton, 1993.

Beaucamp 2007 = J. Beaucamp, Byzantine Egypt and imperial law, in Egypt in the Byzantine World 300-700, ed. R.S. Bagnall, Cambridge, 2007.

Berger 1953 = A. Berger, Encyclopedic Dictionary of Roman Law, Philadelphia, 1953.

Buckland 1966 = W.W. Buckland, A Textbook of Roman Law, Cambridge, 1966 (3rd. ed.).

CAH 1998 = The Cambridge Ancient History, ed. A. Cameron, P. Garnsey, vol. XIII, Cambridge, 1998.

Capogrossi C. 1981 = L. Capogrossi Colognesi, La terra in Roma antica, I (Età arcaica), Rome, 1981.

Capogrossi C. 1988 = L. Capogrossi Colognesi, La città e la sua terra, in A. Momigliano, A. Schiavone (ed.), Storia di Roma, vol. I, Turin, 1988.

Franciosi 1975 = G. Franciosi, Clan gentilizio e strutture monogamiche, I, Naples, 1975.

Kaser 1971 = M. Kaser, Das römische Privatrecht, vol. I, Munich, 1971.

Kaser 1975 = M. Kaser, Das römische Privatrecht, vol. II, Munich, 1975.

Laniado 2002 = A. Laniado, Recherches sur les notables municipaux dans l’empire protobyzantin, Paris, 2002.

Mantovani 2012 = D. Mantovani, Legum multitudo e diritto privato : revisione critica della tesi di Giovanni Rotondi, in J.-L. Ferrary (ed.), Leges publicae. La legge nell'esperienza giuridica romabna, Pavia, 2012, p. 707-767.

Millar 1992 = F. Millar, The Emperor in the Roman World, London, 1992 (2nd. ed.).

OLD = Oxford Latin Dictionary, ed. P.G.W. Glare, Oxford, 1982.

Schmidt-Hofner 2008 = S. Schmidt-Hofner, Regieren und Gestalten. Der Regierungsstil des spätrömischen Kaisers am Beispiel der Gesetzgebung Valentinians I., Munich, 2008.

Sirks 1991 = A.J.B. Sirks, Food for Rome, Amsterdam, 1991.

Sirks 1996 = Ein früh-byzantinisches Szenario für die Amtswechslung in der Sitonie. Die griechischen Papyri aus Pommersfelden (PPG) mit einem Anhang über : die Pommersfeldener Digestenfragmente und die Digestenüberlieferung, ed. A.J.B. Sirks, P.J. Sijpesteijn et al., Munich, 1996.

Sirks 2008 = A.J.B. Sirks, The Colonate in Justinian’s Reign, Journal of Roman Studies XCVIII, 2008, p. 120-143.

Sirks 1994 = A.J.B. Sirks, Sacra, Succession and the lex Voconia, in Latomus, 53, 1994, p. 273-296.

Sirks 2005 = A.J.B. Sirks, Instruction in Late Antiquity, the Law and Theology, in Atti dell’Accademia Romanistica Costantiniana 15, Napoli, 2005, p. 493-503.

Sirks 2007 = A.J.B. Sirks, The Theodosian Code. A Study, Friedrichsdorf, 2007.

Haut de page


1 OLD, s.v. reformo : to alter in appearance, form ; to alter in nature ; to restore to a previous form or condition.

2 Such as nowadays : to bring about a change in behaviour by fines (e.g., fining throwing used chewing gum on the ground), or address social issues by taxation (e.g., taxing the rich to reduce differences in spending power).

3 See Mantovani 2012.

4 The aedilician aediles imposed their edict on market sales. But in view of Varro’s formula’s for the sale of cattle (see his De re rustica 2.2.6 and 2.4.5, e.g.), one may wonder whether the aediles generalised and extended existing cautelary instruments. Yet even then, they would have imposed a prescribed behaviour on all sales of slaves and cattle on the market and forced those who would not have done so before to comply now with this. In that sense it would have been some social engineering.

5 Berger 1953, s.v. sumptus. Such statutes were the leges Aemilia, Fannia, Oppia, Orchia and Iulia Sumptuaria.

6 Buckland 1996, p. 292-293, 319 ; Kaser 1971, p. 318-321, 724.

7 Capogrossi C. 1981, p. 3-12 ; Capogrossi C. 1988, p. 275ff, 267ff.

8 Franciosi 1975, p. 57-68.

9 Buckland 1966, p. 283, 284-285 ; Kaser 1971, p. 106-108, 678-679.

10 See, e.g., Sirks 1994, p. 273-296.

11 Buckland 1966 (3rd. ed.), p. 328 ; Kaser 1971, p. 709-712. It was done by way of the querela inofficiosi testamenti and the descendants could challenge the testament in the first place, then the ascendants, then brothers and sisters. Remarkably enough the wife in manu is not mentioned in the literature, but the sedes materiae, D. 5.2.6 pr. does not exclude her, though dealing specifically with children. This may be due to the disappearance of the marriage in manu in Ulpian’s times.

12 See Buckland 1966, p. 372-373.

13 This senatusconsult was made for mothers not in manu ; otherwise they would have had a claim as agnatic relative of their children.

14 See Sirks 1991, p. 41-42, 61-67.

15 See Millar 1992, p. 3-12, 203-272, 636-652.

16 Schmidt-Hofner 2008, p. 337-344, summarising his research.

17 This is not to suggest that Schmidt-Hofner might be wrong, but for the present argument the question of an effect on social behaviour is the touchstone.

18 Here we see a reform by the introduction of the new rules which established the Tetrarchy and as such this fits the theme of « Réformer la cité », but I would like to explore whether rules were used for wide ranging social changes.

19 Beaucamp 2007, p. 275-285.

20 CJ 5.17.8, of 449.

21 Kaser 1975, p. 177.

22 This does not mean that the interest in sermons, or religion, would not have been genuine. But these served other urges and emotions, which are well imaginable next to the desire for entertainment. In the Middle Ages the Church combined these by the theatrical representations of religious events or quintessences, such as in the play of Elckerlic ; but these are not known for Antiquity.

23 See on the changes in Egypt, our best documented case : Bagnall 1993, passim and part. p. 195, 198-199, 278-289 on Christianity and the Church ; for monasticism p. 289ff.

24 The literature on this is extensive. See for this, as to the connection with the Council of Ephesus, meant to solve the Nestorian conflict, Sirks 2007, p. 81-82, 185.

25 E.g., Donatism had affected entire villages in Africa. See on Donatism and its repression H. Chadwick in CAH 1998, p. 583-586.

26 The first council where this shows is the Council of Sardica of 343, which established rules to enforce the canons upon bishops : see Sirks 2005, p. 500-502.

27 CAH 1998, p. 371-388.

28 Sirks 1996, p. 102.

29 Laniado 2002, p. 134-149.

30 See Sirks 1991, p. 171-172 ; cases in which the naviculariate already had become an obligation pressing on land : the praedia naviculariorum, see ibidem, p. 180-192.

31 See Sirks 2008, p. 124-128 on the interpretation of origo which made this tie possible.

32 Condicionalis in CTh 8.1, 12.7.3, 14.9.2 ; the gynaecearii in CTh 10.20, the fabricenses in CTh 10.22, the murileguli in CTh 10.20, the city workers in CTh 14.7 and 8.

33 L.R. Burg. 37.2 : Quod si pares fuerint honestate persone, consensus perficit nuptias ; sic tamen, ut nuptialis donatio solenniter celebretur ; aliter filii exinde nati legitimorum locum obtinere non poterint, et fit in hac hereditate successioni locus his, qui ab intestato per agnationem, ac si agnatio defuerit, per cognationem forte succedunt.

Haut de page

Pour citer cet article

Référence électronique

Boudewijn Sirks, « Reform and Legislation in the Roman Empire », Mélanges de l'École française de Rome - Antiquité [En ligne], 125-2 | 2013, mis en ligne le 19 décembre 2013, consulté le 29 avril 2017. URL : ; DOI : 10.4000/mefra.1871

Haut de page


Boudewijn Sirks

University of Oxford – boudewijn.sirks[at]

Haut de page

Droits d’auteur

© École française de Rome

Haut de page
  • Logo École française de Rome