CODIFICATIONS IN LATE ANTIQUITY

The Code of Theodosius in 438 and the Corpus luris Civilis marked a new departure in codification.^{[183] [184] When in 42924 and 435,[185] Theodosius declared that his Codex would be known by the imperial name, we are in a different, more centrally directed world even than that of Hermogenian.}

We must distinguish between the applicability to real life of the individual entries in the imperial law codes and the codes as codes. On the face of it, the contents of the collections of imperial constitutions, the Theodosian Code of 438 and the Justinianic Code of 529, re-edited in 534, do rather well as reflec­tors of real life. They are the admittedly truncated responses of emperors to real problems, brought to them by reports or representations from officials, other interest groups, such as town councils, or petitioners, usually funnelled through officialdom in some way.

Because imperial constitutions were responses to representations, which were often self-interested, such as those from the officials in Euphratensis, or from town councils insisting that councillors should carry out their hereditary obligations, there was also an interest on the part of the recipients in seeing that the emperor's will was observed. Of course, some laws were repeated because earlier enactments had been ignored, but in general, the existence of a law implies a problem, a solution, whether temporary or permanent, and some activity on the part of both legislator and recipient.³¹ The contents of the Digest, too, while collated primarily for educational purposes, were also highly relevant to court practice, despite, or perhaps because of, their antiquity.

But the codes, as codes, are rather different. To return to the question

28 T Honoré, “The making of the Theodosian Code” (1986) 103 ZSS (rA) 133; J Harries, “The Roman imperial quaestor from Constantine to Theodosius II” (1988) 78 JRS 148.

29 CTh 15.11.2.

30 CTh 15.11.1.

31 Harries, Law and Empire (n 23) 82-88.

posed above, what are the features of a law code? For present purposes, we may take it that it must exist in written form although, as we have seen, the ancients acknowledged the importance of the unwritten as well. It must have authority, which may be expressed in several forms, and may not carry overt “state” endorsement. That authority must be accepted by those for whom the law code is made. Cicero's imaginary law code in books 2 and 3 of the Laws to which we will return is not such a law code, because Cicero, although a consul and senator, was not empowered to issue legislation binding on the community as a whole, nor did he have the personal standing (or the desire) to impose his code on the res publica.

A law code may resemble a “constitution”, in the modern sense, in that it provides a set of first principles, but does not elaborate in detail; Justinian's Institutes are the prime illustration of this. It may be a written statement of custom or what had been unwritten but accepted before. Alternatively, it may be an attempt to impose system on a proliferation of legal material, as Theodosius and Justinian both hoped to do. If it is to work, the law code must also be accessible to those requiring to consult or to invoke it. Ancient cities with law codes resorted to inscriptions to publicise and preserve their laws. The Code of Gortyn is an example of this;³² so too were the Twelve Tables, in their day. Inscriptional preservation also protected from unauthorised modifications, as circulation in papyrus or vellum or other perishable forms, through copying and private collections, did not. Conversely, while the text might be fixed, there had also to be scope for the law to evolve.

The Theodosian compilers were aware of some of these requirements and did their best to meet them. Detailed instructions were given as to the copying and storing of the Codex and the protection of its text; these were read out to the Roman Senate by the prefect Anicius Glabrio Faustus in December 438.³³ The appearance of its provisions in the, ironically, post-Roman West in such diverse places as a bishop's commentary on Ambrose's De Fide, the Breviarium of Alaric in 506, the so-called Consultatio of some ancient jurist and the Merovingian scriptoria³⁴ suggest that, in the West, the Theodosian Code passed the accessibility test.

But just how relevant, really, was the act of codification to the real life of the

32 J Davies, “Understanding Gortyn: When is a code a code?”, in L Foxhall and A Lewis (eds), Greek Law in Its Political setting (1996) 13.

33 Gest Sen 7.

34 I N Wood, “The code in Merovingian Gaul”, in J Harries and I Wood (eds), The Theodosian Code.

Studies in the Imperial Law of Late Antiquity (1993) 161.

Empire? Take the question of language. Did the citizens of Greek-speaking Antioch or Alexandria - or even Constantinople itself - receive law codes written in Latin with the same orchestrated jubilation as that expressed in the Gesta Senatus at Rome in December 438? One rather doubts it, although there was probably an increase in the learning of Latin in the Greek East in the fourth century and John Matthews has pointed out that Latin remained strong in court circles till well into the sixth century.^[188] Even under the Roman Republic, some enterprising governors of provinces in the Greek East issued their edicts in Greek. Latin was the language of government, among the governors, but Greek was the language of communication with the popula­tions of the Eastern Empire. Even in 438 (let alone 530), the appearance of a Latin law code in Constantinople may have raised a few eyebrows - including perhaps those of Cyrus of Panopolis, who, as Prefect of the Orient in 439-440, would be the first to issue his edicts directly in Greek.^[189] How much stranger would have been the emergence in the 530s of the Corpus Iuris Civilis, with its notion that the law students of Berytus and Constantinople should learn their law in a language, which had no application in daily life - even while Justinian rolled out his Novellae in Greek?

The Latin language was, symbolically and traditionally, the language of power.

Floating on a sea of Greek, the Latin-speaking lawyers may have been prisoners of their culture to a greater extent than we realise. Imbedded in the thinking of both emperors was an assumption that law was both useful (in court) and educative, in the sense that it created better citizens. A long philosophical tradition of law as education stretched back to Plato and other philosophers, via Cicero. The latter's Republic and Laws were extant in Late Antiquity and carried sufficient weight to be points of reference for Augus­tine and others.

Acts of power, presented with dubious accuracy and adorned with philo­sophical musings of doubtful relevance, do not necessarily make for concep­tual or legal coherence. Theodosius' timetable was determined by a non-legal event, the marriage of his daughter, and he failed to follow through his initial three-code design; Justinian was an emperor in a hurry. So should we unques­tioningly have more confidence in Justinian and Tribonian than we might have in their modern equivalents, whose efforts are often greeted with public scepticism. What factors, unacknowledged by their authors, might under­mine the practical application of the late Roman law codes?

First, in Theodosius' mind, at least, there was confusion about what the codification project was for. Was the “first code”, as envisaged in 429, to contain only laws valid now? Or was it to contain also past material for the educational benefit of students of the legal tradition? The provision that the latest dated law invalidated contradictory earlier laws appeared to take care of the question of validity for purposes of citation in the courts and allowed the legal historians and educationalists to have their way. However, it left some questions unanswered. Were later laws always best? How was the volume of material to be controlled if obsolete and superseded constitutions were included because of their antiquarian interest? What should happen if an extant law had fallen into abeyance, but never been formally rescinded?

CTh 1.1.5, of March 429, is the minute of a committee and reflects a compromise between different views as to the purpose of the project. I suspect that the conflict continued in the selection of the material used. There are some cases where an earlier constitution is repealed by a later, for example on the holding of the Maiuma (in Gaza) permitted in 396 and banned in 399,^[198] and the residential status of rowdy monks;^[199] but both are included (the legal educational value of these two examples is unclear). Others are repealed by implication. The first extant law in the collection has Constantine informing the Lusitanians in 322 that “edicts and constitutions”, which lacked a date and consular year of issue, should have no authority; yet this was obviously superseded by the Code itself, by virtue of the rules set out in extract 5 under the same titulus. But we also know of at least one case of an obsolete constitu­tion in effect repealed by omission, as having “fallen into desuetude” (being ideologically unacceptable to a Christian court). This is Julian's notorious law forbidding Christians to teach the pagan classics. The compilers did have the text of the law in front of them, and included the part of it which they deemed valid,^[200] but not the offending clause. This case of repeal by omission is a reminder of the quasi-legislative powers vested in the editors.

Modern users of the Code may know less about the contents than they might like to think. All imperial letters and edicts were “authored” by the emperor, but some may do no more than repeat the report or proposal made, and are therefore more accurately the work of the official or group who requested the law. Many constitutions were sent to multiple recipients; we cannot be sure, in every case, that the named recipient was the actual originator of the issue requiring a legal response. Many contained multiple enactments on one or more topics. Julian's law on teachers had clearly been “buried in perpetual silence”, as Ammianus put it, for some time but it may never have been formally revoked. It made sense, therefore, to repeal the law by ignoring it. What we cannot know is how much else in other laws was being repealed by being ignored.

Perhaps the greatest difficulties facing the compilers of the imperial codices as practical guides to law in action were the related problems of completeness and exclusion. These were (even) worse for the Theodosian Code than for the Corpus luris Civilis, as the latter was the completion of the project, which Theodosius had left unfinished. Like Justinian, Theodosius had envisaged a “digest” of juristic opinions, followed by a complete summary of all law; the last was to be the real “Theodosian Code”. That never happened. Instead, the judges of the Roman Empire were invited, inaccurately, to treat as a complete statement of Roman imperial law, the present Theodosian Code, as modified with additional novellae, for which a process of exchange between the two members of the imperial college was set out.⁴⁸ This exchange failed to function and the absence of any standing commission for the revision of the Code itself meant that the dossier became increasingly unwieldy (and occasionally incomprehensible) over time.

The most serious limitation on the Theodosian Code, at least, was that it was not, and could not be, a complete statement of the law. Because imperial constitutions were so often issued as responses to legal queries, they were themselves a form of commentary or interpretation, not a complete statement of the “law on X”, if such were possible. Yet it would be all too easy for hard- pressed officials to assume that what Constantine to Theodosius II said about, say, treason, was all that there was to say. As I have argued (twice) elsewhere,⁴⁹ the apparent failure of the Theodosian Code to discuss treason in terms of anything other than attacks on the imperial maiestas created a lethal trap for the arrogant and unwary Arvandus, twice Prefect of the Gauls in the 460s. Brought before the Senate by the Provincial Council of the Gauls in 469 and charged with extortion and treason, Arvandus acknowledged publicly (and unnecessarily) that he had dictated a letter to Euric, king of the Visigoths, advising a carve-up of Roman Gaul between the Goths and Burgundians, iure gentium. The spontaneous reaction of the Senate and Board of Judges was to shout out that this was indeed treason. No jurist was cited - the authority in the Digest is Ulpian's De Officio Proconsulis⁵⁰ - because this was something that “everybody” (at least in Rome) knew (although the acclamations were followed up by a volley of legal justifications). Yet correspondence with the

48 CTh 1.1.5; NTh 2.1.3.

49 J Harries, “Sidonius Apollinaris, Rome and the barbarians: a climate of treason?”, in J Drinkwater and H Elton (eds), Fifth Century Gaul: A Crisis of Identity? (1992) 298; J Harries, Sidonius Apollinaris and the Fall of Rome (1994).

50 D 48.4.1 (Ulpian, Duties of the Procunsul 7).

enemy is nowhere cited in the extant Theodosian Code, which excluded all imperial constitutions - but not jurists - outside itself. Instead, it discusses offences “resembling” maiestas, such as counterfeiting imperial coinage or the illegal manufacture of purple cloth; procedural issues, such as torture; and penalties. Little wonder that officials, perhaps with little formal legal education, were unaware that the emperor's was not the last legal word.

Such were some of the factors separating law codes from life. Clearly, law codes had their merits in making law accessible. But setting law in stone, as it were, inhibited the law's responsiveness to social and legal change. The very existence of a “code” created stresses in the system. Within the late-Roman codes were preserved the outlines of what might be termed quasi- or crypto­codes from previous centuries. In these certain types of law were included and arranged in a fixed order. The edict is the most obvious of these and its long history, both in its own right and as the framework for voluminous juristic civil law commentary, ensured its continued usefulness. It was even possible, after its “codification” to incorporate other actions, to be heard “extra ordinem.^[201] As, in my view, the concept of ordo, and thus of extra ordinem, is the product of the conservatism of juristic discourse, I avoid its use.

There was, I suggest, a problem with public criminal law, which goes back to Augustus' statutes on the indicia publica, discussed above. By the second century, the “public crimes” list had become canonical. But society's view of what is wrong or criminal is not static. There had to be ways of changing the canon, or its contents, without breaking the link with tradition. Various means were employed, none ofwhich amounted to a coherent programme of reform. Senatorial decrees, imperial decisions and, perhaps, court decisions assimi­lated other offences to one or other public statute. When the late-Roman law codes approached the question of public law, they did so via sections headed “on” a given canonical statute, and included the assimilated offences as well as the original ones. The end of the use of juries in the quaestio procedure, based at Rome, and the universal adoption all over the Empire of the cognitio conducted by one judge allowed a further adaptation, namely the hearing of cases which were not covered by the edict, and offences which were not part of the Augustan “ordo”. In cases of doubt, a magistrate could resort to his policing powers and exercise summary justice, without resort to statute.

Should we grant Augustus' law the status of a quasi-codification of the publica indicia? One sure indication is the coincidental appearance of a commentary, in this case that of Ateius Capito (also undated), who became consul in AD 5 (Gellius Noctes Atticae 4, 14; 10.6). Its existence is known, not from the Digest, but from Aulus Gellius' collection of interesting snippets compiled at Rome in the second century. Unlike the second-century inter­preters, Capito did not write a separate treatise on the subject but entitled the ninth book of his Coniectanea (“Bits and Pieces”), De Publicis ludiciis. How far this was a “legal” commentary is unclear: according to Gellius, himself an antiquarian with legal interests, Capito praised tribunician and aedilician edicts drawn from the distant past and aimed against the rudeness of a Claudia in the third century BC and a rowdy Mancinus in the second.

Whether or not Capito was more than an anecdotal antiquarian (the selec­tion we have now would be influenced by Gellius' own predilections) jurists had not commented seriously on criminal law before, and they were not to do so again until the subject took off in second-century Rome, the city of the antiquarians Pomponius, Gaius and Aulus Gellius, and of the arch-codifier, Salvius Julianus. Like Ateius Capito, Lucius Volusius Maecianus was a politi­cian, who would be Prefect of Egypt; he wrote fourteen books on the publica iudicia. Less detailed, and perhaps for that reason more extensively cited by the hard-pressed redactors of the Digest, was the three-book effort on the same subject of Claudius Venuleius Saturninus. But if the publica iudicia were suddenly trendy, it was also becoming clear that their separate status created procedural anomalies. Public trials, for example, required an accuser, while police proceedings did not. The debate on theft and the Twelve Tables “recorded” (or made up) by Aulus Gellius^[202] argues that some activities were “criminal” in a social sense, although in a legal sense they counted as delicts, to be pursued by private procedures. By the time of Ulpian (if we are not in the presence of an interpolation), theft was proceeded against “crimi­naliter”, which could mean “by (public) accusation”. For Aemilius Macer, writing under Severus Alexander, the separate survival of the publica iudicia seems to have been something of an embarrassment: “not all the courts in which an accusation can be brought are public courts, but only those which derive from the statutes on the publica iudicia, for example the Lex Julia on maiestas (etc)”.^[203] For another Severan jurist, the statutes had outlived their usefulness; Callistratus, who had no known ties with Rome, abandoned the public/private schema altogether.^[204]

But the compilers of the Digest did not. The second of the so-called Libri Terribiles (Book 48) faithfully followed the canon, with a series of chapters “ad” a statute, as did also the codifications of imperial law. More intrigu- ingly, they may not have worked too hard on it. Under the Severans, the publica indicia, with the exception of adultery, which received wider and more detailed coverage, were dealt with in two-book manuals by Macer and Marcianus, who are both cited thirty or more times. But there was a scholarly obligation to look further back and in Antonine Rome they would have found Maecianus' fourteen books and Saturninus' three. The citation of Saturninus twelve times and the presumably more demanding Maecianus, on points of details, a mere four times, may suggest perhaps a certain laziness (or lack of time) on the part of the compilers of Justinian's definitive statement on criminal law. An alternative (slightly irrelevant) speculation is possible. Something of the career of Lucius Volusius Maecianus was known at Rome in the fourth century, to the author(s) of the Historia Augusta, who dated him, correctly, to the reigns of Antoninus Pius and Marcus Aurelius.^[205] So the infrequency of the citations is not due to his obscurity. On the other hand, the literati of fourth-century Rome were addicted to biographies (rather than long histories) and epitomes, to which we may add other forms of potted literature. Could it be that, in sixth-century Constantinople, as in fourth­century Rome, Maecianus was known, not through his complete text, but legal anthologies?

E.