The Making of the Text

John Matthews

The Theodosian Code was promulgated in the West at a meeting of the Roman senate held on 25 December 438.

The situation revealed by the law is one of some interest, evidence of the popularity and usefulness of the Code; it would appear that demand already outstripped the rate of production, and even competent judicial authorities were forced to seek pirate copies. This may be inferred from the text of the law itself: ‘sacrilegii poena constringi tam cognitionale officium quam eos, qui nostris minime paruerunt constitutis’ - where the italicised words refer to officials and their departments. The Gesta Senatus were meticulous and emphatic on the question of maintaining the authenticity of the text of the Code and the authority of the constitutionarii-, it was no doubt the latter who ensured the preservation of this confirmation of their rights, by appending the Gesta and the imperial law to copies of the Code subsequently produced by them, one of which descended to the manuscript in the Ambrosian Library.

*

The Gesta Senatus begin by naming the consuls, the Emperor Theodosius (for the sixteenth time) and Anicius Acilius Glabrio Faustus. The full name of Faustus, ‘tertio expraefecto urbi, praefectus praetorio et consul ordinarius’, also heads those of the three high officials present at the meeting: the others, mentioned in order of seniority, are Fl. Paulus, v.c. et inl(ustris) urbis praefectus, and lunius Pomponius Publianus, vir spectabilis vicarius urbis aeternae. Then follow the ‘proceres amplissimique ordo senatus’. The senators, having assembled, had some discussion, the nature of which is not recorded, and then there joined the meeting by invitation (‘ingressique ex praecepto’) the constitutionarii Anastasius and Martinus; these are the Fl. Anastasius and Hilarius Martinus to whom, still at their work five years later, Valentinian’s law of 23 December 443 was addressed. The constitutionarii were not senators, and it was necessary by ancient convention that they should be formally invited to attend the deliberations.

Anicius Acilius Glabrio Faustus then addressed the meeting. He began by reminding his audience of the circumstances. He recalled his attendance the year before at the wedding of the young Valentinian III to Theodosius’ daughter Licinia Eudoxia. He does not name them, nor the place and date of the wedding, which was in fact at Constantinople on 29 October 437.^[17] Faustus was one of the legati sent by the Roman senate to attend the occasion: a colleague on the embassy was Rufius Antonius Agrypnius Volusianus, praefectus urbi in 417-18 and praefectus praetorio in 428-9.^[18] In earlier years Volusianus was a famous correspondent of Augustine, and he is mentioned as a newly-appointed urban prefect in the poem of Rutilius Namatianus, De Reditu Suo\ his illness and late baptism, and his death on the day of Epiphany (6 January) 438 while on an embassy to Constantinople, are described by the biographer of his niece, the younger Melania, who had travelled from the Holy Land to visit him.

After the completion of the wedding ceremonies (continued Faustus), Theodosius had announced his intention to present to the world a compendium of imperial law, to be called after himself: and this project Valentinian, ‘loyal as a colleague, loving as a son’, had approved: ‘quam rem aeternus princeps dominus noster Valentinianus devotione socii, affectu filii conprobavit’ (Gesta Senatus, 2). Needless to say, the project was already completed; when it was begun, in 429, Valentinian had been just ten years old.

At this point Faustus paused, while his words were greeted with acclamations of approval of their eloquence: ‘nove diserte, vere diserte.’ When these were over (unlike the more elaborately orchestrated series of acclamations later in the meeting, there is no indication of the number of repetitions) he continued by describing how Theodosius had summoned to his presence Faustus himself and the praefectus praetorio of the East at that time, and had personally given to each of them - or, more accurately, had ‘ordered to be conveyed to them from his divine hand’ - a copy of the new codex: we may indulge ourselves for a moment by imagining the scene, as an imperial functionary advances, takes the splendidly bound copies of the Code from the emperor and solemnly presents one to each of the prefects in turn. It may be that it was on this very occasion that Faustus received promotion to the praetorian prefecture of Italy, Illyricum and Africa, indeed that this was his first official act in that capacity. An inscription from Aricia in Latium, on which the promotion is mentioned, records it as an expression of the judgment of each part of the Empire: ‘utriusque inperii iudicii(s) sublimitato.’ We may well see this, from a western perspective, as a promotion to a western prefecture taking place at the eastern court.^[20] So Faustus had received the Codex, given with the direction (‘praeceptione’) of each emperor, and he completed his initial statement by acknowledging the presence in the senate of the constitutionarii and proposing the reading out of the law (he actually says ‘leges’) by which the emperors had brought all this about.

Again follow acclamations of approval - ‘aequum est, placet, placet’ - and Faustus, presumably taking the copy from the constitutionarii, proceeded to read out the law of 26 March 429, addressed to the senate of Constantinople, by which the entire project was initiated.^[21] The law is cited by Faustus ‘ex codice Theodosiano, libro primo, sub titulo “de constitutionibus principum et edictis” ’; it is 1.1.5 in the edition of Mommsen and Meyer and is the fundamental evidence for the manner in which the Theodosian Code was compiled, and for its purpose.

The Theodosian Code was to be a successor to and to complement the earlier (late third-century) codes of Gregorius and Hermogenianus, and would follow the same pattern of composition. It would include all imperial constitutions possessing the force of edicts or of general application issued by Constantine and his successors down to the present day: ‘omnes constitutiones, quas Constantinus inclitus et post eum divi principes nosque tulimus edictorum viribus aut sacra generalitate subnixas.’ The laws were to be arranged by titles (or subjects), with the proviso that extended texts containing material relevant to more than one subject were to be divided, and the separate parts classified under the titles to which they individually applied; and, whether intact or as the separate parts of divided laws, they were to be entered under their titles with their dates, and in the sequence in which they were originally issued. In this way the actual ordering of the material would make clear which constitutions on any particular subject were the later and therefore the more valid: ‘ipsius etiam compositione operis validiora esse, quae sunt posteriora, monstrante.’ Extraneous matter not relevant to the point of substance in a law was to be omitted - a stringent, and in a way disappointing requirement, given the vigorous rhetoric habitually used to enhance the force of late Roman legislation:^[22] and the Code was to be complete, even to the extent that earlier laws clearly superseded by later were nevertheless to be included.^[23] It would be more convenient, the emperor admitted, if only currently relevant law were to be codified, but the Theodosian Code, like its predecessors, was to be as complete as it could be made. It was to be a work for specialist scholars and practitioners (‘diligentiores’), whose professional interest (‘scholastica intentio’) would wish to know even of laws valid only for their own time and now obsolete.

The constitution of 429 went on to nominate an editorial panel of nine persons with powers to co-opt experts (‘eruditissimum quemque adhibituros confidimus’), and it concluded with a provision, in effect a limiting clause, on the relevance in either part of the Empire of laws issued in the other: such law was only to be valid if it was officially directed to, and accepted and published in, the part in which it was received.^[26]

With these instructions, we may suppose that the editorial panel of nine, with their co-opted assistants, had enough guidance to set about their task. However, the emperors returned to the subject in a law issued at Constantinople on 20 December 435, which again, with some repetition, defines editorial procedures (CTh 1.1.6). Again we find the instructions to collect all laws issued by Constantine and his successors up to the present day, to arrange by titles, dividing laws where necessary, and to omit material not relevant to the substantive point at issue.

It has sometimes been argued that the new arrangements represent a substantial amendment, or even a cancellation, of those of 429. Mommsen thought that the law of 435 indicated an abandonment of an earlier plan that had turned out unsuccessful, a view expressed also by Gaudemet, and by Jolowicz, who wrote of ‘a new commission with different instructions’ set up by the law of 435.^[27] Wolfgang Kunkel supposed that ‘the commission appointed by the emperor [sc. in 429]... evidently produced nothing’, and that a ‘second commission called into being six years later’ completed in two years a modified version of the initial project of Theodosius.^[28] A closely similar view was expressed by A.H.M. Jones.^[29] Archi, whose discussion of the relevant evidence is the most subtle and perceptive known to me, still writes of an ‘old’ and a ‘new’ programme, and of ‘saving what could be saved of the project of 429’.^[30]

I am convinced that these interpretations involve a misreading of the evidence.^[31] The constitution of 26 March 429 stated clearly that the project of a critical digest that would contain only currently relevant law was one set aside for the future. It was indeed contingent upon the publication of the Theodosian to add to the Gregorian and Hermogenian Codes (‘ex his autem tribus codicibus... noster erit alius’, etc.), but was not in editorial terms a part of the current project. Jolowicz’s observation, again echoed by Jones, that ‘the project of including juristic writings was abandoned [sc. by the law of 435]’ is also misconceived. It was only later that it was envisaged that the writings of the jurists would be included, to produce a definitive, critical compilation of Roman law, both ‘leges’ and ‘ius’. This too was part of the next phase of the project. The constitution of 429 gives no guidance on that subject, nor was it ever meant to.

It is also clear that the process of compilation and editing set out in 429 would in the best of circumstances, and with the best will in the world, take a long time, as the editors and their co-opted assistants applied themselves to recovering imperial constitutions from what, at least on my view of their methods (see below), were widely scattered sources, in both eastern and western Empires, as well as from whatever central and provincial archives were available: and, when this stage was complete, to a detailed process of editing. We ought not to be surprised if such a project were not brought to completion in all its stages in the 6³/4 years from March 429 to December 435. Conversely, it is beyond belief that, as the views of Mommsen and others just quoted would imply, the project should have been rescued from failure, revived and completed, even in a modified form, in less than two years between December 435 and its publication as a complete text in October 437. We may add that, by the mere fact of its being read out, the senatorial meeting of 25 December 438 was invited to look back to the law of 429 as initiating a process brought to completion without interruption. Indeed, unless, as is unlikely, something has gone wrong with the text of the Gesta Senatus, the law of 435 was not even read to the senate.¹⁷

It is more convincing to see the law of 435 as in part reiterating, in part refining the principles laid down in 429, at a specific point in the editorial process^ that is to say, at the moment when the task of collecting the laws was complete and that of organising and editing the material could begin. This conclusion is supported by certain differences between the laws of 429 and 435, which I would suggest are better seen as refinements of an existing method than as definitions of anew one:¹⁸

(i)    The constitution of 429 had defined the laws to be collected as those ‘with edictal force or [i.e. in other ways] of general application’: ‘constitutiones... edictorum viribus aut sacra generalitate subnixas.’ The constitution of 435 repeats this provision in the phrase ‘omnes

¹⁷        Cf. Honoré, 166; Sirks, Chapter 2 §7 below and op. cit. (n. 11 above), at 277 on the (non)-reading of CTh 1.1.6.

¹⁸        Cf. also Honoré, 165f.; Sirks, Chapter 2 §7 below: ‘It seems as if the design of 435 concerning the editing of the constitutions is more detailed than that of 429, or even replaces it.’ I agree with the first of these statements (insofar as they are intended as alternatives); and further, ‘[the] design did not abrogate that of 429, but elaborated upon it’. My view of what happened between 429 and 435 differs however from that of Sirks, who sees a shift from a more scholarly emphasis in 429 to a forensic/administrative one in 435, which ‘found its expression in the exclusion of obsolete rules’ (p. 58; cf. above, n. 7). I regard the constitution of 435 as an internal stage in the process of editing rather than as reflecting any change of circumstances or policy. See below, n. 52 for a striking example of editorial completeness. edictales generalesque constitutiones’, but adds the apparently radical, one might even think contradictory, amendment that laws valid or published in particular provinces or places were also to be included: ‘vel in certis provinciis seu locis valere aut proponi iussae’ (1.1.6). In formal, juristic terms, this is not as surprising as it looks, once we set it in the context of what is meant by a law of ‘general· application. The best and most relevant definition of ‘generalitas’ is given in Codex Justinianus 1.14.3, part of a long constitution addressed to the senate on 6 or 7 November 426, of which this particular paragraph is either for some reason not excerpted, or does not survive, in the Theodosian Code.¹⁹ According to this text, which must have been known to the editors of the Code, ‘generalitas’ consists of pronouncements sent to the senate in the form of imperial orationes or issued by the emperors as edicta, whatever the specific situation that might have occasioned them; and, more broadly, of imperial pronouncements made public by provincial governors which had relevance to other cases. ‘Generalitas’ is also defined by limitation, to exclude rescripts issued to individual litigants,²⁰ and ‘interlocutiones’, or legal rulings made in the course of specific proceedings without the intention that they should be applied more generally.

All this seems in formal terms comprehensible, and certainly should have been so to the editors of the Theodosian Code. The difficulty was that the formal definition does not comprehend the varied nature of the actual material which the compilers turned up. This was by no means primarily ‘juristic’ in character, nor on a commonsense view defined by its general relevance, but covered any subject, large or small, upon which the emperors had made pronouncements. This was remarked upon by Gibbon as he ‘gratefully remembered’ how, in studying the Code with the commentary of James Godefroy, he had found it far more than a book of law in any narrow sense: ‘I used it (and much I used it) as a work of history rather than of Jurisprudence: but in every light it may be considered as a full and capacious repository of the political state of the Empire in the fourth and fifth centuries.’²¹

If Gibbon’s remark is welcome support for the historian’s rights of settlement in the territory of Jurisprudence, it also highlights the difficulties of selection and judgment faced by the editors of the Code. Even if they began their work by applying the most formal understanding of generalitas - that is, by excluding all rescripts and

¹⁹        For the law, sometimes known as the ‘law of citations’, cf. Mommsen’s ed., Prolegomena, p. CCCI; Seeck, Regesten, p. 352 (dated 7 November). It survives in seven citations from CTh and five from CJust. See esp. on these laws, in relation to the issue of generalitas, Sirks (n. 11 above), at 286-91.

²⁰        A brief statement of Arcadius restricting the validity of‘rescripta ad consultationem emissa vel emittenda’ to the proceedings (‘negotia’) in respect of which they were issued survives as CTh 1.2.11; 6 December 398. See Harries above, p.13.

²¹        Memoirs of My Life (ed. G.A. Bonnard, 1966), p. 147. interlocutiones - they must often have found their material very difficult to classify. They will have found many cases where principles which clearly could have possessed force as general laws were enshrined in pronouncements addressed to individual officials and posted in particular places; it cannot always have been clear whether any general force was intended, particularly in cases where such pronouncements had been superseded by later ones addressed, it might be, to superior officials. The editors would appreciate that even in their most specific pronouncements the emperors often gave general reasons for acting as they did; it was simply not the case that points of general application were only ever embodied in laws of a formally general character. They must also have realised that many of the laws recovered were versions addressed to particular officials, of laws originally more general in scope which had not been recovered. They would know too that the distribution of late Roman legislation was sometimes a very complicated process. As laws passed down the line from emperor to praetorian prefect to provincial governor to local community, they underwent changes in form - as from imperial epistula to prefect’s or governor’s edictum - and acquired commentary by intermediate authorities; a law brought to the attention of the compilers of the Code might in principle belong to any of its various phases, and might not be directly from an emperor at all. Codex Theodosianus 7.13.11, for instance, issued on 15 May 382 ‘at Tyre the metropolis’ and posted at Beirut, cannot derive from an imperial letter but must be part of an edict of the praetorian prefect of the Orient sent on to a provincial governor, and there are other such cases.^[32] The Gregorian and Hermogenian Codes, cited as precedents in the law of 429, further confused the issue, since these earlier collections, though much more ‘juristic’ in character than the texts which now lay before the editors, consisted almost entirely of rescripts. These, as we have seen, were together with interlocutiones specifically excluded from the category ofgeneralitas.^[33]

Formal assumptions made by the editors about the character of generalitas in the law must repeatedly have clashed in common sense with the specific character, and often the sheer oddity, of the situations in which their ever-conscientious emperors had expressed themselves. Considering such a law as that which appears as Codex Theodosianus 7.1.13, instructing that horses be washed downstream out of sight of military camps, so that drinking water was not polluted by mud and sweat (and worse), nor the eyes of beholders by the sight of naked grooms splashing about in the river, the editors might have inferred a general principle relating to public health and propriety. The law stated a comprehensible general benefit to the public, and one could readily see why, in its necessarily particular circumstances, it should always be observed. A more elusive principle lay in such a provision as that expressed in a law of 355 addressed to the praetorian prefect: ‘Henceforth, whenever cured pork, or perhaps fresh meat, is issued to soldiers stationed in Africa, We command that only the hoofs and tips of the noses of the animals shall be cut off and removed, so that all the pork shall not be separated but shall serve in the issue of the food supplies’ (7.4.2; transl. Pharr). The generalitas of this regulation lay, not in its wide reference to all sorts of situations, but in the fact that it was formally stated to be applicable to any similar situation in the future.^[34] One could readily accept that it was wrong for the recipients of exported meat to be deprived of part of their rations by cunning suppliers who had the carcases divided before they were dispatched, and sometimes ‘lost’ joints of meat to their own advantage, but this principle is implied rather than stated in the text of the constitution. In any case, it hardly belongs to the juristic heights of Roman law. The better-known law of 397, posted at Rome in the Forum of Trajan and on pain of exile and confiscation of property banning the wearing of Oriental and Germanic trousers in the streets of the venerable city, derives its generalitas from the categorical nature of the ban and from the fact of its address to the people, rather than from the universality of the situation envisaged; this could only, by definition, affect the city of Rome.^[35]

We have to imagine our editors, confronted with cases like this, constantly attempting to reconcile juristic notions of generalitas with the actual content of the laws which they discovered. Were they really to suppose that the washing of horses, the snouts and hoofs of slaughtered nigs or the wearing of trousers at Rome, were issues involving generalitas, just because the emperors had pronounced upon them?

In view of such considerations I suggest that we might see in the phrase ‘vel in certis provinciis seu locis valere aut proponi iussae’ not an amendment of their original instructions but a definition made to help the editors of the material as they contemplated their task, and as the book of law on which they were engaged looked progressively more unlike any other book of law which they knew. It responds to queries raised by the editors as they began their work of excerpting and classifying the material they had collected, this material having proved much more varied in character than was allowed for in their original instructions. The tendency of the definition would be to increase the scope of what was included in the Code by subjecting the editors’ material, often against the apparent implications of its actual content, to the most formal definition of generalitas. Without its guidance, the editors might well have been tempted to omit the snouts of slaughtered pigs.²⁶

(ii)    Editorial considerations also explain a second innovation in the constitution of 435. The editors were now empowered not merely, as in 429, to abbreviate laws by omitting extraneous matter but, in order to ensure the clarity of the published texts, to expand where necessary, to alter what was unclear and to emend what did not make sense: ‘et adiciendi necessaria et demutandi ambigua et emendandi incongrua.’²⁷ Again, this looks like a response to a request from the editors, based upon inspection of the material they had collected, that they be given these powers. It was prudent to settle this point, since even the most obscure words of the emperors issued from their divine mouth (which is, indeed, sometimes called an ‘oracle’). As Symmachus learned as praefectus urbi in 384, to question the emperor’s judgment in making appointments was an act of sacrilege, which might cost the offender ten pounds, and his officium five pounds of gold in fines.²⁸ To amend the emperor’s utterances might be no less hazardous than to question his judgment; it was as well to be careful.

It i^{s a}S^ain worth emphasis that the problems sketched here confronted the editors n e outset of their work, when the full range of imperial conscientiousness had become th^em but before they had had much time to digest it.

ed t S^onoy®’ ^SSRA 103 (1986), 165f. on these powers and their restrained use by the v ^{1 S}. ^r₽iⁿ (n. 9 above), at 622 n. 7 regards this clarification of the meaning of texts that of429 ^{worcolor=black face="Century Schoolbook">[37]}

*

This interpretation of the procedures employed by the editors of the Theodosian Code is of course connected with, indeed is one way of viewing, the question of the actual, physical sources from which they recovered the texts of their laws. This question is discussed by Boudewijn Sirks in Chapter 2 of this book, with a preference for central archives - in particular copy-books (Registerbiicher) in which the emperors entered copies of the legislation they issued - over the provincial archives favoured by Seeck.³¹ A consequence of this theory is that the first stage of editing I have been describing, based, in this case, merely on the recovery of imperial copy-books and such provincial archives as are required to explain the 200 proposita datings relied upon by Seeck as indicating provincial origin, would have been a relatively compact and painless operation, and one that can hardly without good excuse have taken 6³/4 years; other causes for what now appears as a delay between 429 and 435, such as a change of policy or emphasis, have to be postulated.³²

My own view is that we should not work with too restrictive a conception of the sources available to the compilers of the Code. It would be absurd to deny in principle the use of imperial and provincial archives if these existed, whether in the form of copy- or day-books, select commentarii, or in whatever form they may have been kept. On the other hand, the idea that central archives in any sort of continuous form were the main or essential source for the compilers of the Code fails sufficiently to explain the discrepancies between laws, and between different versions of the same laws which can be found, and which seem most easily explained on the assumption that their texts were derived from different sources. It would clearly be dangerous to assume that all such discrepancies indicate provincial or local origin and then to assign all laws that do not show them to central sources, since not all discrepancies may be apparent to us. We also have to account for the omission from the Theodosian Code of laws which are known to have existed from literary sources such as Eusebius’ Vita Constantini, independent collections like the Fragmenta Vaticana and the Sirmondian Constitutions, from occasional inscriptions, and from references to lost earlier legislation made in surviving texts of the Code itself.³³ It is of course possible to postulate whatever incompleteness in or damage to archives as is required to explain the omissions in the

³¹        Sirks, Chapter 2 §4 below.

³²        Sirks, Chapter 2 §7 below. For my reservations on these, cf. above, nn. 7, 18. See below, nn. 53, 62 for indications of recovery from local sources; proposita datings are not always required to establish this.

³³        E.g. to laws of Constantine; CTh 13.5.26, 9.3; 14.3.12; 16.10.2; to Gratian’s law on temple property, 16.10.20. Wenger, op. cit. (n. 12 above), 537, emphasises the ‘Unvollstandigkeit der kaiserlichen Archive’.

Code that are found or may be suggested,³⁴ but this can be a rather tautological procedure. In general it seems more likely, not that the 'missing’ laws were found but omitted from the Code, as suggested by Sirks, but that they were never found.³⁵ Indeed, if there were substantial central archives at the disposal of the compilers, one might have expected far more laws to have been found than seems to have been the case. The total of 2,700 ‘basic texts’ postulated by Sirks (plus whatever, on his view, was omitted by the editors) does not seem an especially impressive haul of legislation if the originals survived in central archival sources approaching anything like completeness over the whole period 311-437.³⁶ It is much more impressive if central sources were very incomplete and the laws had to be recovered from a range of diverse provenances. Furthermore, the argument that laws were recovered from officially maintained copy-books, whether central or provincial, leaves out of account one of the most obvious and important ways in which law is known to have been preserved in the Roman Empire; by the copying down by interested parties of publicly posted texts. I will briefly return to this below.

I believe, therefore, that no single or narrow range of possibilities relating only to officially preserved versions of imperial legislation explains the discrepancies and anomalies that do in fact occur in the published texts of the Code. The answer to the question of the sources of the Theodosian Code lies not at one or two, but at many points in the very wide area between the central records perhaps consulted in the case of some recent laws issued in the East, and the diverse and scattered locations, especially in the West,³⁷ in which the laws recovered were originally received. Given the wide variety of individual anomalies that may be found, in what follows I shall proceed by the illustration of possibilities rather than by general analysis, and will begin with two cases of relatively recent eastern legislation where it seems clear from the edited protocol, either that

³⁴        As a matter of interest and indirect relevance, B. A. van Groningen, A Family-archive from Tebtunis (1950), at 44-62 and 85-108 provides a fascinating account of a dispute over (i.e. avoidance of) the municipal duty of maintaining the public archives; cf. esp. (among many other difficulties), p. 56 on ‘[documents] missing for the greatest part the beginning, some also, as is usual, in a damaged condition... [some] lost, being tom and worn by age, others also have been partly damaged and several have been eaten away at the top because the places are hot’. Cf. E. Posner, Archives in the Ancient World (1972), atl51f.

³⁵        Sirks, Chapter 2 §7 and his n. 74 below. Sirks’ view is linked with a different understanding from mine of the phrase ‘ad vim sanctionis non pertinentia’ as a criterion for editorial exclusion in the constitution of 429, and in general about what was going on between 429 and 435.

³⁶        Sirks, Appendix 1; to the basic 2,700 texts Sirks adds [y], the number of constitutions collected but omitted by the editors of the Code. It will be clear that we take different views of [y], cf. nn. 7, 52.

³⁷        The predominance of western legislation collected in the Code, pointed out by Seeck, Regesten 2, does require explanation (cf. n. 62 below).

the compilers had found various copies or versions of laws, which they gave in a single text while indicating the alternative sources that had come to their notice, or that they found a single copy already annotated in the fashion shown. Both laws are discussed in much the same sense by Seeck.^[38]

(1)           1.8.1: IMPP. HONORIUS ET THEODOSIUS AA. FLORENTIO MAGISTRO MILITUM. Viro illustri quaestore Eustathio suggerente, etc.... DAT. ID. OCT. CONSTANTINOPOLI HONORIO X ET THEODOSIO VI AA. CONSUL. SCRIPTA EODEM EXEMPLO SAPRICIO MAGISTRO MILITUM, HELIONI MAGISTRO OFFICIORUM ET EUSTATHIO QUAESTORI [15 October 415].

(2)           11.28.9: IDEM AA. ANTHEMIO PP.... DAT V ID. APRIL. CONSTA(NTI)NOP(OLI), CONSTANTIO ET CONSTANTE CONSS. DE EADEM RE SCRIBTUM EDICTUM AD POPULUM: AD MARCIANUM COM(ITEM) S(ACRARUM) L(ARGITIONUM): MUSELLIO PRAEPOSITO SACRI CUBICULI DE TITULIS AD DOMUM SACRAM PERTINENTIBUS: AD RECTORES PROVINCIARUM: ET DE METALLARIIS EDICTUM AD POPULUM PER PROVINCIAS ILLYRICI ET AD RECTORES PROVINCIARUM [9 Aprii 414].^[39]

The first of these laws concerns what Honoré (p. 77 below) calls ‘an item of official infighting’, the restoration of certain military appointments (praepositurae) to the laterculum minus or ‘lesser register’ maintained by the quaestor, from the magistri militum who had taken them over. For this purpose copies would have to be sent ‘eodem exemplo’ to all the relevant officials, including the magister officiorum, who had general supervision of all court arrangements; any one of these copies might in principle have been recovered. The law is also a modest bureaucratic success for the influential quaestor Eustathius who ‘suggested’ it. It does not follow that it was in his department that it was found, but a central archive, whether his or that of the magister militum, might be the best interpretation of the text and style of annotation of this recent law.^[40] In the case of 11.28.9 it might be a valid deduction from the reference to the publication of a part of the text as an 'edictum ad populum per provincias Illyrici et ad rectores provinciarum’, that the document was recovered as annotated at an early stage from the archive of a central official, possibly that of the praetorian prefect who had responsibility for its further diffusion. Had this 'edict concerning mineworkers’ been recovered only in the version received and published in Illyricum, it would not have been defined in this way. Like the 'edict to the people’, and the section to Musellius referring to 'accounts pertaining to the sacred imperial household’, it would have needed amendment and commentary in order to be intelligible by itself.

We might similarly view as deriving from central sources certain very late inclusions in the Code, of laws issued in the period between the revised instructions issued to the editorial panel in December 435 and the completion of the text in late 437. Since the last dated laws in the Theodosian Code, from summer 436 and spring 437, were addressed to successive praetorian prefects of the Orient, Isidore and Darius, we might assume that these laws - one of those to Isidore, of 4 June 436, being edited in six extracts, three of them in succession (12.1.189-191)^[41] - were made available to the editors of the Code from sources close to the centre of power. In a further example from this late period, however, three edited laws deriving from a single pronounce­ment of 3 April 436 were addressed to the praetorian prefect Isidore, two of the extracts but not the third preserving the annotation that the same law was also issued to the praetorian prefect of Illyricum:

(3)  8.4.30: IDEM AA. ISIDORO PPO. POST ALIA: EODEM EXEMPLO EUBULO PPO ILLYRICI... DAT. Ill NON. APRIL. CONST(ANTINO)P(OLD ISIDORO ET SENATORE CONSS.

(4)  12.1.187: IDEM AA. ISIDORO PPO... ET CETERA. DAT. Ill NON. APRIL. CONSTANT(INO)P(OLI) ISIDORO ET SENATORE CONSS. EODEM EXEMPLO EUBULO PPO ILLYRICI.

The interest in this group of laws lies partly in the fact that the prefect of Illyricum mentioned, Eubulus, was a member of the editorial panels both of 429 and of 435 but had evidently been promoted from his work on the Code to take up the prefecture;^[42] he, if anyone, was au fait with the proceedings of the panel and in a good position to ensure that the commission received contemporary legislation in its best and most authentic state. Despite this, the third version of this law (12.1.188), largely duplicates, but also differs in detail from, 8.4.30, and in its protocol makes no mention of the prefect of Illyricum. It seems clear that in this very recent case the compilers of the Code recovered and worked with not one, but at least two copies of the law.

If this is so even for the most recent eastern legislation, the situation for the more distant past, and especially for the western Empire, is much more complicated. Here we find anomalies that are very difficult to reconcile with the notion of a single archive maintained, in whatever form, at the centre of power.^[43] By what process of editorial recovery, for example, do we possess the text of a single law of Constantine 'accepted’ at Hispalis (Seville) on 18 April 336, apparently a full four years after it was issued at Constantinople?^[44]

(5) 3.5.6: IDEM A. AD TIBERIANUM VICARIUM HISPANIARUM... DAT. ID. IUL. CONSTANTdNO)P(OLI). ACCEPTA ղϲ K. MAI. HISPALI NEPOTIANO ET FACUNDO CONSS.

A four-year gap between issue and acceptance is excessive on any assumptions of bureaucratic delays or difficulties in communication between East and West in the time of Constantine, and it may be that this is a special case, to be solved by a more radical approach than any so far thought of to problems of textual transmission or fourth-century prosopography. A convincing historical context can however be found for discrepancies in a law of Valens addressed to the comes sacrarum largitionum, which survives in two fragments of what - given the diversity of subject-matter in the two fragments - was clearly a much longer law issued at Noviodunum on 3 or 5 July 369. One fragment and not the other of this law is (with other variations in the protocol) recorded as 'accepted’ at Marcianopolis on 18 July:

(6) 10.16.2: IDEM AA. ET GRATIANUS AD ARCHELAUM COM(ITEM) OR(IENTIS) [sc. COM. S. L.]... DAT. Ill NON. IUL. NOVIODUNO VAL(ENTINI)ANO N(O)B(ILISSIMO) P(UERO) ET VICTORE CONSS.

(7) 10.21.1: IMP. VAL(ENTINI)ANUS ETVALENSAA. ARCHELAO COM.S.L....DAT.V NON. IUL. NEVIODUNI; ACC. XV KAL. AUG. MARCIANOPOLI VAL(ENTINI)ANO NOB. P. ET VICTORE CONSS.

It is not difficult to work out what has happened here. In 369 Valens made Marcianopolis his base for his third campaigning season against the Goths, advancing to Noviodunum on the lower Danube for the conduct of hostilities. It looks from 10.21.1 as if the emperor left behind his court administration at Marcianopolis, and addressed this law to his comes sacrarum largitionum there from his campaign headquar­ters.^[45] The text clearly derives not from the original version issued by the emperor and copied into some central file, but from that received by the comes sacrarum largitionum. The different protocol of 10.16.2 (with the correct inclusion of the name of Gratian), the transmitted discrepancy in the date, and the error in the office of the recipient, make it clear that this part of the law was recovered from a different source, which again is unlikely to be a central archive; had it been so, the compilers would have found both laws together as part of the same text, without the discrepancies between them that we find in the published versions. The compilers have here recovered two partial, already separated, versions of the same law but because of the differences in dating and protocol did not realise that they were dealing with a single pronouncement.

In another example, subtly different versions are found of a law addressed to the vicarius of Africa on the evasion of municipal duties under the assumed guise of high office. One law is recorded as ‘accepted’ at Carthage on 12 December 338, the other as ‘posted’ somewhere unspecified (not at Carthage, since the date is earlier) on 27 November and ‘accepted’ at Timgad on 15 December:^[46]

(8)12.1.24: IDEM A. ACONIO CATULLINO VIC. AFRIC. Quicumque fugientes obsequia curiarum affectaverint adumbratae nomina dignitatis, etsi eos spes falsi honoris inluserit, XXX argenti libras inferre cogantur. ACC. KARTHAG. PRID. ID. DEC. URSO ET POLEMIO CONSS.

(9) 6.22.2: imp. constantius a. aconio vic. afric.... cui addimus, ut quicumque fugientes obsequia curiarum umbram et nomina adfectaverint dignitatum, tricenas libras argenti inferre cogantur, etc. PP. V KAL. DEC., ACC. XVII IAN. THAMUGADI URSO ET POLEMIO CONSS.

The second of these versions, in which the recipient’s name is also transmitted incompletely, perhaps represents a copy sent on by the vicarius to the city of Timgad, improved by someone in the uicarius’ office who fancied his hand at re-touching the style (perhaps he thought that he was simplifying it). In any event, it is clear that the compilers of the Code recovered two versions of the law and not a single, authoritative copy from some central archive. It is also of interest that CTh 12.1.25 & 26, issued respectively on 28 October and 1 November (the latter also to Catullinus), nevertheless follow 12.1.24 in the published text of the Code. The editors must have realised that, although the recorded date of acceptance at Carthage of 12.1.24 was later, it was issued earlier than its successors. That this was an editorial inference is clear, since the date and place of issue of the law are not recorded at all.^[47]

At 16.1.4 the Theodosian Code preserves one of the most widely-reported of late Roman laws, that issued by Valentinian II at Milan on 23 January 386 permitting the free assembly of Arian congregations. The circumstances which led up to and the consequences which followed the publication of the law are described by various writers, including Ambrose bishop of Milan {Ep. 20) and Augustine, who was in the city at the time as professor of rhetoric {Conf. 9.7(15)), and the church historians Rufinus (11.16) and Sozomen (7.13.5ff.). From Rufinus and Sozomen, and from Gaudentius bishop of Brescia, we hear about the latter’s parishioner the magister memoriae Benivolus, who resigned his office rather than draft the legislation:^[48]

(10) 16.1.4; IMPPP. VAL(ENTINI)ANUS, THEOD(OSIUS) ET ARCAD(IUS) AAA. AD EUSIGNIUM PPO.... [scituris] his, qui sibi tantum existimant colligendi copiam contributam, quod, si turbulentum quippiam contra nostrae tranquillitatis praeceptum faciendum esse temptaverint, ut seditionis auctores pacisque turbatae ecclesiae, etiam maiestatis capite ac sanguine sint supplicia luituri [manente nihilo minus eos supplicio, etc.]... DAT. X KAL. FEB. MED(IOLANO) HONORIO NOB.P. ET EUODIO CONSS.^[49]

That the circumstances of this law concern Milan and the relations of the imperial court, influenced by the Arianism of the empress and her supporters, with the Catholic community of the city and its bishop is beyond any doubt. It is puzzling, therefore, to find at CTh 16.4.1 a single sentence of the law cited almost (not exactly) verbatim and with the same date as 16.1.4, but presented as if issued at Constantinople:

(11)            16.4.1; IMPPP. VAL(ENTINI)ANUS, THEOD(OSIUS) ET ARCAD(IUS) AAA. EUSIGNIO PPO. His, qui sibi tantummodo existimant colligendi copiam contributam, si turbulentum quippiam contra nostrae tranquillitatis praeceptum faciendum esse temptaverint, ut seditionis auctores pacisque turbatae ecclesiae, maiestatis capite ac sanguine sint supplicia luituri. DAT. X KAL. FEB. CONSTANT?. HONORIO NOB.P. ET EUODIO V.C. CONSS.

The textual differences in this version are slight but from an editorial point of view significant. Apart from the different place of issue, ‘Eusignio’ is read for ‘ad Eusignium’,^[50] and the consul Euodius is given his rank ‘v(ir) c(larissimus)’; Euodius was the consul of the western usurper Magnus Maximus, and his name is here recorded, as in many laws, in the year in which he was recognised by Theodosius.^{Schoolbook">[51]} The minor discrepancies of phraseology do not suggest that the shorter text was edited from the longer, or either text from a common original; compared with 16.1.4, 16.4.1 omits the words ‘quod’ and ‘etiam’ but on the other hand reads the longer ‘tantummodo’ where 16.1.4 has merely ‘tantum’. Although it is true that it concerns the most general aspect of 16.1.4, it is hard to see how 16.4.1 ever became connected with the eastern Empire; for present purposes, it is relevant only to observe that the editors of the Theodosian Code must have acquired these two laws from different sources. Had they extrapolated 16.4.1 from 16.1.4 they would not have incorporated the discrepancies in the protocol that I have described.^[52]

In similar fashion, the only likely explanation of a law addressed to the consularis of Campania or the consularis of Picenum is that the compilers recovered two copies of a law originally issued more widely, perhaps to all provincial governors, to all governors of Italian provinces, or to a praetorian prefect with the instructions to forward it to them:^[53]

(12) 12.1.71: IDEM AA. ET GR(ATI)ANUS A. AD AMPHILOCHIUM CONS. CAMPANIAE SIVE AD SOFRONIUM CONS. PICENI... DAT. Ill NON. MAI. TREV(IRIS) VAL(ENTINI)ANO ET VALENTE III AA. CONSS. [5 May 370].

The same explanation might hold for a law addressed to the proconsul and the vicarius of Africa but beginning with a phrase in the singular, and so clearly addressed to them separately:

(13) 1.12.6; IDEM AA. VICTORIO PROCONSULIAFRICAE ET DOMINATORI VICARIO AFRICAE. Apparitioni tuae et legatorum quadringentos de his dumtaxat, quos rei publicae membra non querentur sibi esse detractos, etc. [including also the phrases ‘arbitrii tui’ and ‘officii tui’]... DAT. XII K. IUN. MED(IOLANO) HONORIO A. IIII ET EUTYCHIANO CONSUL. [21 May 398].^[54]

In another instance of a single law preserved in two versions addressed to the same official, the Theodosian Code and an independent source, the Collatio Legum Mosaicarum et Romanarum, derive from copies of a law posted in different places and at different dates in the city of Rome:

(14) 9.7.6; IDEM AAA. ORIENTIO VICARIO URBI ROMAE.... PP. IN FORO TRAIANI VIII ID. AUG. VALENTINIANO A. IIII ET NEOTERIO CONSS. [6 August 390].

(15) Collatio 5.3 (Riccobono, fira², p. 557): IMPP(P). valentinianus THEODOSIUS ET ARCADIUS AUGGG. AD ORIENTIUM VICARIUM URBIS ROMAE... PROP. PR. ID. MAIAS ROMAE IN ATRIO MINERVAE [14 May, SC. 390].

It would in principle have been possible for the compilers of the Code to have recovered either of these, or indeed other, copies of the law, which the vicarius of Rome would have had posted in public places in the city in order (see below) that they might be read and copied by interested parties. An additional benefit in this case is that the version in the Collatio (the subject is male prostitution) preserves the original rhetoric that the editors of the Theodosian Code, following their editorial instructions concerning 'matters not relevant to the force of sanction’ Cad vim sanctionis non pertinentia’), omitted.⁵⁵ It is enough for legal purposes that burning alive be prescribed as the sanction. It is not necessary from a legal or editorial point of view also to evoke the improving effect of this procedure upon onlookers, vividly expressed in the longer version of this law.

Particularly intriguing, to return in conclusion to purely technical considerations, is the varied collection of dates and recipients presented by the group of laws in the Code which are textually related to - though not derived from - the law preserved to us as the the sixth of the Sirmondian Constitutions:⁵⁶

(16) Sirm. 6: IMPP. THEODOSIUS ET VALENTINIANUS CAESAR AMATIO V.I. PRAEF. PRTR. GALL.... DATA VIIIDUS IULIAS AQUILEIAE D.N. THEODOSIO A. XI ET VALENTINIANO CONSS. [9 July 425].

This version of the law preserved in the Sirmondian Constitutions restores legal privileges and benefits to the clergy in the aftermath of the usurpation of lohannes at Ravenna, making particular reference to the treatment of heretical bishops in Gaul. The prefect is also instructed to convey the imperial commands to the knowledge Cnotitia’) of the provinces under his jurisdiction. It is evidently from such further instructions sent on to governors and officials, in this case in Italy and Africa, that the compilers of the Theodosian Code derived their corresponding material:

(17) (i) 16.2.47: IDEM A. ET CAES. BASSO C(OMITI) R(EI) P(RIVATAE)... DAT. VIII

ID. OCTOB. AQUIL. D.N. THEOD. A. XI ET VAL(ENTINI)ANO CONSS. [8 October].

(ii)            16.5.62: IMP. THEOD. A. ET VAL(ENTINI)ANO CAES. AD FAUSTUM

⁵⁵        Cf. above, n. 7 for the meaning of this phrase.

⁵⁶        Seeck, Regesten, 5.

P(RAEFECTUM) U(RBI)... DAT. XVI KAL. AUG. THEOD. A. XI ET VAL(ENTINI)ANO CAES. CONSS. [17 July].

(iii)       16.5.64: IDEM A. ET CAES. BASSO C.R.P. POST ALIA... ET CETERA. DAT. VIII ID. AUG. AQUIL. D.N. THEOD. A. XI ET VALENTINIANO CAES. CONSS. [6 August].

(iv)       16.2.46: IMPP. THEOD. ET VALENTINIANUS CAES. GEORGIO PROC. AFRIC. POST ALIA... ET CETERA. DAT. PRID. NON. IUL. AQUIL. D.N. THEOD. A. XI ET VAL(ENTINI)ANO CAES. CONSS. [6 July].

In addition, 16.5.63 (cf. 16.5.62 and 64 above) seems likely to be connected in some way with the same pronouncement:

(v)        16.5.63: IDEM A. ET CAES. GEORGIO PROC. AFRIC.... ET CETERA. DAT. PRID. NON. AUG. AQUIL. D.N. THEODO. A. XI ET VAL(ENTINI)ANO CAES. CONSS. [4 August].

The fragments of the law preserved in the Theodosian Code clearly derive from versions issued to officials working under the praetorian prefect of Italy (the comes rei privatae) or as part of separate jurisdictions (the urban prefect and the proconsul of Africa) in the area of the Italian prefecture;⁵⁷ there is no textual dependency between these versions and the separate tradition relating to Gaul preserved in the Sirmondian Constitutions. Nor, apparently, was the corresponding version of the full text of the law, which we can assume to have been addressed to the praetorian prefect of Italy, available to the compilers of the Code. These derived their texts from partial versions addressed on five different transmitted dates to three different officials operating within the Italian prefecture. Yet this is a recent law, which would surely have been preserved in full in imperial copy-books, had these been available.

*

Such cases as these pose endless problems in the detailed interpretation of the Theodosian Code, many of which are discussed, with due regard for the uncertainties, in the pages of Seeck’s Regesten, in The Prosopography of the Later Roman Empire, and in T.D. Barnes’ The New Empire of Diocletian and Constantine. My purpose is rather different; it is to argue, following in this respect Seeck and A.H.M. Jones, that the compilers of the Theodosian Code cannot have worked with systematic access to central archives or registers of legal sources; had that been the case, there would be more laws, fewer discrepancies between different versions of the same laws, fewer indications too of the circumstances in which laws were received, read out, and posted in public places. This returns us to the very important point mentioned

⁵⁷        The prefect of Rome and proconsul of Africa stood outside the hierarchy headed by the praetorian prefect, and reported direct to the emperor; cf. A.H.M. Jones, The Later Roman Empire I, pp. 374f. with n. 22. earlier. It was by hearing the law announced in public places and reading it when posted on notice-boards, usually in ephemeral form - only a very small proportion of imperial legislation achieved the permanent status of being engraved on bronze or marble tablets⁵⁸ - that the populations of the Empire were expected to have knowledge of it, and legal authorities to take note by making copies for themselves. I need here only illustrate this very common practice of copying down imperial legislation: from criticism of the Emperor Caligula because he imposed new tax laws and had them engraved in very small letters and in an awkward corner, so that they could not be copied,⁵⁹ to the quotation by Eusebius, in a Greek translation of a Latin original, of Galerius’ edict of toleration; and, later, from the copy engraved (again in Latin) at Tyre, of Maximinus’ rescript against the Christians, a document also known from two inscribed copies from Asia Minor.⁶⁰ Of the two versions of the so-called ‘Edict of Milan’ of Constantine and Licinius offering toleration and rights of ownership of property to the Christians, one was cited by Lactantius from the copy posted at Nicomedia, which is where he was living at the time (De Mort. Pers. 48, cf. Eusebius, HE 10.5.2-14); it was in that city, as Lactantius also records, that an angry Christian had torn down an edict against the Christians from the board where it was posted, declaring sarcastically that victories over Goths and Sarmatians were being promulgated (De Mort. Pers. 13). The text of one of the most important of all Roman legal enactments, the so-called ‘Constitutio Antoniniana’ of AD 212, survives because it formed part of a dossier of legal documents assembled by an individual who wished to use it to support his title to residence at Alexandria, or some other question of social status or immunity to which the ‘Constitutio’ was relevant.⁶¹

⁵⁸        E.g. for engraving on bronze, CTh 12.5.2 (given as ‘pp. Karthagine’, so not recovered from an engraved version), ‘quod ut perpetua observatione firmetur, legem hanc incisam aeneis tabulis iussimus publican’, cf. 14.4.4; for engraving on bronze or painting on wooden tablets or linen posters, 11.27.1, 'aereis tabulis vel cerussatis aut linteis mappis scribta per omnes civitates Italiae proponatur lex’; for posting in public places, 16.5.37, ‘proposito programmate celeberrimis in locis volumus anteferri et gesta, quibus est huiuscemodi allegatio inserta, subnecti’. The instruction to display legislation in public places and legibly is a common provision recorded on inscriptions, cf. esp. Callie Williamson, ‘Monuments of Bronze: Roman legal documents on bronze tablets’, Classical Antiquity 6.1 (1987), 160-83.

⁵⁹        Suetonius, Caligula 41.1: ‘minutissimis litteris et angustissimo loco, uti ne cui describere liceret.’

⁶⁰        Eusebius, HE 9.7.3. For this and the epigraphic evidence, adding a new Latin inscr. from Colbasa (on the road Antalya-Burdur) to the Greek inscr. already known from Arykanda in Lycia, see the fundamental and fully-documented discussion by Stephen Mitchell, ‘Maximinus and the Christians in ad 312: a new Latin inscription’, JRS 78 (1988), 105-24. For Tyre as the metropolis of Phoenicia, see above, p. 27.

⁶¹        P. Giss. 40; cf. F. M. Heichelheim, ‘The text of the Constitutio Antoniniana and three other decrees of the Emperor Caracalla contained in Papyrus Gissensis 40’, JEA 26 (1940), 10-22, esp. 21-2. See also Callie Williamson (n. 58 above), 165f. for the copying of military diplomata.

Between the extremes of centrally maintained archives and ephemeral display on the notice-boards of the cities of the Later Roman Empire, there existed an array of opportunities for recovery from intermediate sources, particularly from law schools (Berytus, where fourteen laws in the Theodosian Code are recorded as ‘posted’, is an obvious example, though it is in a way surprising that there were not more),^[55] and other places where lawyers practised and made working collections of imperial enactments, but also from individuals and groups who had an interest in copying and preserving the law on particular subjects; and one may add to these the possibility countenanced by Seeck and A.H.M. Jones, of the recovery of legislation from the family papers of individuals, primarily aristocratic, who had kept them after their departure from office.^[56] On the assumption, however valid on any of these grounds, that the recovery of so large a proportion of western laws was not achieved without some actual visits, at least to major centres in the West at which laws might have been collected and made available for later recovery, it is worth reflection (hat Che period in which the recovery took place, 429-435, was the last.n vhich such a project, involving travel in an increasingly insecure western Mediter­ranean, could easily have been envisaged. In conclusion it is therefore worth pausing for a few moments on the political context in which the project was conceived and implemented.

On 23 October 425 an era of co-operation between eastern and western governments had been inaugurated by the instaJation by the generals of Theodosius of the child emperor v_alentinian ill. and by the restoration to Ravenna of his mother Galla Piacidia. for the past two years a refugee at the court of Constantinople.^[57] The marriage of Valentinian and Licinia Eudoxia in October 437 was the culmination and symbol of the unity of the imperial house achieved in this period. If the process of accumulation of the law that was achieved in the years 429-435 is a tribute to the actual possibilities of travel and commu­nication in the Mediterranean world achieved by the political unity of the imperial house, the publication of the Code, coinciding with the marriage of Valentinian and Eudoxia, was a corresponding symbol of this political unity.

But we have only to glance at the immediate future to see what a narrowly-achieved symbol it was. In 430, just a few months after the Theodosian Code received its impetus from the imperial government, the Hippo of St Augustine came under Vandal siege. In a process that coincides almost exactly with the stages of editing and publication of the Code, the Vandals were conceded settlement in north Africa by a treaty concluded at Hippo on 11 February 435, and in October 439 took possession of Carthage. In Gaul, Gothic settlement in Aquitania, and later Burgundian settlement in the Rhone valley, began to break the hold of the government of Ravenna on all but the Mediterranean seaboard of the north-western provinces of the Empire. The preservation of the Theodosian Code in just these areas of Germanic occupation of Gaul reflects a very different historical phase from that in which the accumulation of the primary material for the Code had taken place. It was a period of political uncertainty in the western Mediterranean world, in which the actual compilation of the Theodosian Code, seen, in part at least, as a physical process of travel and access to widely scattered sources, can begin to seem not only a supremely optimistic, but even a modestly heroic, activity. The Theodosian Code, the first fully official attempt since the publication of the Twelve Tables by a Roman government to collect its own legislation,^[58] is one of the last expressions, as well as one of the last symbols, of Roman imperial unity.^[59]

2.