Where did the Theodosian compilers take their texts from?
p. 153-164
Résumés
Dans son ouvrage Laying down the Law. A Study of the Theodosian Code, J. F. Matthews a défendu la thèse, déjà avancée par O. Seeck, qu’en fait, tous les textes du Code Justinien antérieurs à environ 380 avaient pour origine des édits provinciaux puisque publier une loi était nécessaire pour en assurer la validité. Notre propos est ici de montrer que tel n’était pas le cas. Tout d’abord, ce n’était pas là pratique courante sous la République romaine ou le Haut-Empire, ensuite parce que les définitions des règles générales dans CJ 1, 14, 2 et 3, ne l’impliquent pas. Il s’en suit que les textes des lettres, envoyés par l’Empereur et conservés dans les manuscrits des archives impériales étaient source suffisante pour les compilateurs du Code Théodosien. Recourir à d’autres sources n’était nécessaire que lorsque les archives impériales étaient insuffisantes dans ce domaine, mais cela fut beaucoup moins fréquent que ne l’avance Matthews. Cela ne peut qu’avoir une incidence sur le degré de fiabilité de la datation des textes.
In his Laying down the Law. A Study of the Theodosian Code, J. F. Matthews has defended the thesis, already submitted by O. Seeck, that basically all texts in the Justinian Code before ca. 380 derive from provincial edicts, since publication was necessary for the validity of a law. It is argued here, that was not the case: firstly, since that was not the practice in the Roman Republic and Early Empire, secondly, because the definitions of general rules in CJ 1, 14, 2 and 3 do not imply that. This means that the texts of the letters, sent by the emperor and preserved in the copybooks in the imperial archives, sufficed for the Theodosian compilers. Recourse to other sources was only necessary where the imperial archives in this respect failed, but that was much less the case than assumed by Matthews. This bears upon the reliability of the dating of the texts.
Texte intégral
1Where many ancient historians occupy themselves with texts from the Theodosian Code, it is fitting to consider some basic problems regarding these texts, particularly since a decade ago John Matthews’ Laying Down the Law. A Study of the Theodosian Code has brought these problems into the limelight again.
2Those questions relate to the texts of the Theodosian Code and their sources. The Theodosian compilers got the order to gather all laws issued by Constantine and his successors and to distribute these over books and titles, according to subject-matter. The texts we dispose of are often contradictory regarding the issuing emperor or the date of issue. About 85% merely has data with a date as subscription1. The remaining 15% show more, mentioning accepta with a date or even proposita with a date. If one assumes that all laws had to be published in every province in order to obtain force, one must assume that these laws were collected there, and further, that if the date of a data does not fit another data, one must assume that the laws were marked on reception with a date of acceptance and on publication with a date of publication, so that the date in question must be the date of a lost accepta or proposita; and that is exactly what Seeck did and Matthews does. The consequence of this, namely, that all subscriptions with only data become suspicious in this way and any dating illusory2, they do not accept, however. The solution is found in the idea that the constitutions with only data would have come from imperial archives, where copies of sent letters were kept (and in that case the date would be secure — but then the question of validity rises); the others from archives of the addressees. Matthews assumes that only for the later period, beginning between 380-395, the compilers used basically the eastern archives; for the anterior period, the compilers had to rely far more on other sources: collections made in the provinces, provincial archives3. This position was already taken by Seeck4. Yet the question is, may we split the constitutions with only data and one date into two groups, one reliable, the other not, as regards the date, since there are many such constitutions dating from before that time? And what does the date mean in the first case, if it is not the date of publication? And what about the validity of the rule? Did it depend on a date whether a rule was valid enough to be included or not?
3For our understanding of the way the texts in the Theodosian Code were collected, selected and put in chronological order these questions are important. The compilers had to collect rules, which were, or had been, of general validity: what, then, was the basic text used by the compilers? and connected to this: to which sources could or did the compilers turn or have to turn to? These questions depend, again, on three other preliminary questions:
what was the normative, authoritative text of a general rule (in whatever form issued),
when did such a text gain force of law, and
how did one get knowledge of it? For that we have to consider the legislative process in the Empire.
4According to Matthews, who also discusses first the lawgiving process as it was during the Late Republic and Early Empire, the basis of the emperor’s lawgiving authority was the lex de imperio, which made that the emperor disposed of imperium and thus of legislative authority5. It is true that the emperor could exercise this as well as judicial authority, but here I would define it more precisely and state, that his laws were laws in the form of an edict, and that all other rules issued by an emperor were equated to edicts and by that to leges rogatae. The reason for this is first that imperium as well as tribunicia potestas (which both powers the emperor had) implied the power to issue edicts, and second CJ 1, 14, 3, where leges generales are equated to edicts if they conform to certain formal characteristics. We see indeed that for the later period no lex de imperio is mentioned, but the tribunicia potestas still is, until it has become customary constitutional law6. As to the imperial rescripts, Matthews does not mention a basis for their authority7. We have suggested elsewhere that here the authority lay in the emperor as pontifex maximus, which gave him the ius respondendi, the right to answer decisively on legal questions put before him8. That is conform to the general nature of the rescripts which deal with legal matters: they never enter into the actual case and its facts. It required here a judge to issue the judgment.
5There is another form of imperial legislative activity: the letters sent to high officials with the order to publish by edict what the emperor had decided9. Apparently Matthews assumes (on the basis of Gaius 1, 5), that these were laws too, which is right, but he does not analyse how Gaius could say so or why that was the case in the Later Empire.
6What was the normative text in the above cases? With law, lex, generally a normative rule is meant. It will depend on the issuing instance what formal qualities such a rule had (is it an edict, a lex rogata, or a decision, taken in a case and used in other cases, etc.). Here we see a remarkable thing: though there are second century collections of imperial constitutions and judgments10, there are no collections of leges rogatae. Yet this is, when closer considered, not strange. A lex might sanction an act, but it would only have effect, if transformed into a formula in the praetorian edict. Thus we see that it is the text of the formula which is cited and commented upon, as in the case of the actio de pauperie, which ex lege duodecim tabularum descendit11. Likewise for penal laws (actio popularis), whereas with the luxury laws it was slightly different: they were applied by the censor in his regimen morum. It is therefore not a surprise that we do not find something like a «Corpus of Roman Laws». What we, also know for the Republic and Early Empire is, that acts, edicts and particularly the Pretorian Edict were published upon issue12. But orality was and remained the basis of validity, since in early times laws were established by question and answer in the comitia, edicts were, as the word says, spoken, and testaments were oral (nuncupative), with witnesses to remember the contents. The normative text was therefore the text spoken and they were at once valid.
7Likewise procedures were oral, and after the formulae were introduced, the judgment still was oral and again at once valid. One might surmise that with the increase in literacy written versions set the oral versions aside, but that is not likely, since we see still in the late third and in the early sixth century decided, that, notwithstanding the contents of a written testament, the orally uttered will of the testator was presumed and prevailed13. The same will have been the case with the edict, which literally means: «what is pronounced/ordered» (edictum).
8With rescripts the immediacy of the process also points to orality as basis. A rescript had to be handed in by the supplicant himself or a close relative, in case of a village by a village-member. He then had to wait one or some days for the answer, which was written underneath the plea. This seems to be a procedure by writing, but the original form was probably the morning salute of one’s patron with the oral question for assistance or information. Had it been changed into a true written procedure, with the written answer as decisive point, the supplicant might have stayed home and merely sent his plea. With greater towns and higher-ranking persons a letter sufficed, but here this rather was deference to their position, not forcing them to go all the way of a mean client. The basis remained the salutatio with oral conference and immediate reply14.
9That being so, the next question is of course how could one be certain what the spoken text was? For testaments the tabulae, sealed with the seals of the witnesses, was used. For edicts, a publication was custom, and with leges rogatae the text was drawn up before the law was proposed. With senatus consults, which later gained force of law, which were also orally accepted, the text was that of the protocol, as stocked in the archives. Publication was not necessary for the validity of the rule, but it was custom to make the text in some way public.
10As with laws we sometimes see the jurists cite the text of a rescript literally, sometimes paraphrase it. The citation or paraphrase gained the same force as the literal text, since the writings of the jurist were citable in court, as we know from the Lex citandi of 42615. But was it this law, which gave paraphrases such a force? Or did they have already force, being it not necessary to cite the rescript exactly? It is the latter, since the Lex citandi deals with the problem of reliable texts of the jurists’ writings.
11One might surmise that in those days people were not yet fixated on precise texts of laws etc. being content with the formulation of the general purport of a rule, but that is not the case. Too often the jurists refer to the exact wording of a law (lex Aquilia) or cite the very words of a rescript, and the later collections of rescripts and laws testify to this. That a paraphrase was accepted was only because of the reputation of the jurist in question. Partly, for the older laws, this question would have been secondary since the attention was focused on the actio and there we see that the exact wording counted and that hair-splitting exegesis was applied.
12In the course of time in the Empire some matters changed. The emperors issued mandates, letters, delivered speeches in the senate, sent letters to high officials. There were considered to be equal to leges rogatae, laws, formally approved by the people’s assembly. The questions again are: what exactly was considered to be the normative text here? At which moment did the rule become binding? How did people get to know the content of the rule? It might be submitted that they obtained force only when issued as edicts by the provincial governors and not before. In that case a provincial edict in all provinces would be an indispensable requirement, since a provincial governor could only issue edicts for his own province. The alternative is, that the letter itself was already considered equal to an edict, on basis of criteria later, in 426, defined in CJ 1, 14, 3. Was a provincial edict then necessary? Perhaps in order to fulfil the custom of publication.
13These questions touch directly the Theodosian Code, since almost all its texts derive from letters to high officials of the state.
14The eastern emperor Theodosius II ordered in 429 a committee to gather first all constitutions, which had the force of an edict (edictorum uiribus) or were considered to have general validity ordered by the emperor (sacra generalitate)16. The latter requirement removed edicts of the prefects, provincial lieutenants etc., which could, in themselves, also be general rules17. In a later constitution, of 435, Theodosius spoke of omnes edictales generalesque constitutiones uel in certis prouinciis seu locis ualere aut proponi iussae, which are to be placed in chronological order (just the date did not suffice)18. Was this a category, not covered by the description in 429? That is, did the order of 429 only comprise of laws, meant for the entire empire, whereas the order of 435 supplemented this by including in the compilation laws, of general application, but restricted to certain geographical areas? Or did it cover the same category and is the period uel in certis prouinciis seu locis ualere aut proponi iussae merely an indication of the way laws were issued? This latter interpretation is not completely valid. It is clear that in 435 general constitutions of restricted geographical application were to be included in the compilation as well. Yet, the last addition, ualere aut proponi iussae, may well have been applicable to both categories. It would mean that a constitution might be ordered to be valid, or to be posted in a province.
15To a certain extent we are lucky here, since the contemporaries of the Theodosian compilers left us with texts about what was a general norm and what not, although, on the other hand, such statements have to be interpreted as well and may pose problems in their turn. They were drawn up in the west, probably in the same law as the Lex citandi, but there is no reason why they did not represent views, shared in the east. As to what was a normative text and when did it gain force, CJ 1, 14, 3, pr. and 1 state that a rule has uim edicti, the force of an edict (and by that is a general rule), if a) it is called an edict, or b) if it has been made public everywhere by edict of the governors (per omnes populos iudicum programmate diuulgari) or c) if the emperors have expressly stated in the text that it should apply for future similar cases; or d) if it is ordered to extend to all (ad omnes iussa est pertinere, which does not require publication). Matthews assumes that the bulk of the texts in the Theodosian Code, which are letters to high officials, were indeed published by way of a governor’s edict in every provincial capital19, which means he considers them to be of category b). Yet, is that true? How do we know that, the texts being stripped of the parts in which that would have been indicated? For the mentioned categories a), c) and d) publication apparently was not mandatory and so non-publication was not fatal for the validity. For category b) it apparently sufficed for validity, if it had been published everywhere; calling it an edict, e.g., was evidently not necessary. Since it would have been rather cumbersome to check this, probably it sufficed that such a general publication had been ordered, and the proponi iussae in CTh I, 1, 6 appears to confirm this. Matthews, however, assumes that all laws in the Theodosian Code had originally carried such publication orders20. As said, we cannot not know this since the texts were edited, but we do know that it was not necessary for categories a), c) and d), and probably for category b) also if only it had been ordered. If they fulfilled one of the other requirements, they were already valid general laws. For example: CTh V, 13, 1: uniuersi cognoscant... (cat. d), CTh VI, 31, 1: per omnes prouincias edictum generale misimus... (cat. a, publication is not necessary), CTh X, 10, 31: edictali lege sancimus... (cat. a); CTh XI, 1, 34, a rule for African landowners:... quattuor mensum ab edicti publicati die... (here publication in the region of Africa served merely to set the date for complying with the rule, the rule itself already being in force); CTh VI, 23, 4 ends with inlustris auctoritas tua statuta nostra clementiae edictis propositis ad omnium notitiam faciet peruenire. Here it is clear that publication merely served to make the text known to all decurions, not that it was needed for validity; but it probably also already sufficed to have the text fall into cat. b).
16If the addressee wanted to publish it – but he did not have to for its validity – it was to inform the public, but a single publication would have sufficed. This bears upon the question to the normative text as well. It means that the letter or edict, issued by the emperor, was normative, since it gained already force by the fulfilment of the formal conditions. So it simply depended on the text itself.
17So nowhere we find that publication was a requirement for validity of a law and if non-publication had no consequence, we may not be sure that it always happened. However, Matthews maintains that a law did depend for its validity (ualere) on its promulgation and display in the communities to which it was sent21. He rather ambiguously speaks of «the impact of Roman legislation owed its force, etc. », but we must distinguish between the validity of a law in itself and its (subsequent) actual impact. Was a person bound to a law which was not yet promulgated in his hometown? Matthews cites a rabbi of the late third century, according to whom one was not punishable as long as promulgation had not taken place22. This position would mean that a law would acquire force (validity) in the Empire on different moments, at potentially in great temporal distance.
18We have already concluded on basis of the law of 426 and the preceding legislative custom of the Romans that publication was not a requirement for validity, but we like to approach the question raised by Matthews also from this other angle. For that we have to distinguish between validity of a law at the moment of issue, before its publication, and publication as requirement for its validity. In the first case the law would be binding, although not yet known, in the second case only when known, and here an objective knowledge would be required (that is, when one reasonably could have learned of it, for example, by going to the forum of one’s town). However, it is the classical and postclassical rule that ignorance whatsoever of the law is not an excuse: constitutiones principum nec ignorare quemquam nec dissimulare permittimus (CJ 1, 18, 12 of 391). Only persons who enjoyed to some extent in law protection on account of their (socially) weaker position, could use the excuse of ignorantia iuris: minors, women, and even rustici, people (farmers) on the land outside of the towns23.
19Now if in the time of the Theodosian compilers publication would have been mandatory, there would have been a problem here as regards the date of publication, since every general law had then to be posted in every province. That certainly could not have happened everywhere at the same time. Also, if a law was already valid at the moment of issue but ignorance before its publication would excuse, validity would mean nothing; and it would further lead to the awkward conclusion, that, as in the first case, in different provinces different regimes would exist regarding ignorance of a certain law. Yet there is no exception of this duty recorded for the case that a law was not promulgated in somebody’s domicile, or, not unthinkable in view of the excuse for being away on official business (rei publicae causa abesse), that one was away on such an occasion, when the law was published.
20Neither do we see a possible solution for this problem in the form of fixing a uniform starting date of validity after the posting, as is usual today24, and the assumption of Seeck, that the compilers, being aware of this problem, therefore took the date of issuing (data) as the date of general validity, is inconsistent25. The plan of 429 merely prescribes the mention of consulate and date in connection with the question of lex posterior derogat legi priori, and in 435 the order merely added to this that the order of editing should also be chronological. There is no hint of plural dates for one constitution. This may only explained if we assume that the moment of the start of validity was not a problem for the Roman government, and this could only be the case if this moment was the moment the emperor put his name or signing to the draft, or pronounced the text, as it was before with the edicts. This thus also follows from the use of the words constitutiones edictales, constitutions which contained the word edict or edictal.
21All this is in conformity with what we know of law-giving in the Republic and Early Empire, as exposed before, and there is no reason to assume that a change had taken place. We saw with the republican laws that they only became effective if incorporated in, e.g., the praetorian edict. This may have changed in the Principate, in the cognitional procedure (after 342 universal, as a result of the abolition of the formula system). Here it would have sufficed to mention a new law in the libellus. Edicts and senatus consults were also valid at once, but only the first were published, the latter merely recorded in the archives. Speeches held by the emperor in the Senate also gained force of law, but there as well validity must have been the immediate effect of the speech, probably already the effect of sending it, since CJ 1, 14, 3 says uel missa ad uenerabilem coetum, but in any case not the effect of the later archiving or publication (if any). Rescripts were, though publicly posted, only meant for the requesting person. That publication of a law was not mandatory in the Later Roman Empire fits this situation. If the emperor issued an edict, the moment he recited the words or approved of the final version of the text, it became law26. This also explains why around 426 the need was felt to lay down that a letter in which the emperor ordered to issues provincial edicts with a new rule, was already considered an edict if only it contained the word edictalis. There must have been doubts concerning its normative power in itself. One might also argue, that the letter was not normative, but that the edicts were, yet that the imperial officials (quaestors) considered the thought of a norm entering into force thoughout the entire empire on potentially very divergent moments unbearable and that for that reason they decided, that on the moment the text of the letter was approved of, the rule obtained validity. It seems to me that the basis of this assumption poses already a divergence of the foregoing praxis, but in any case its adherents must then accept a) that publication by provincial edict was not a requirement anymore and b) the fact that the letter itself was considered an edict, which was not published in itself in order to obtain force (since that is not required in CJ 1, 14, 3). Also the extension of the edictal character (and by that of general binding) in CJ 1, 14, 3 to rules, formulated as generally binding, can only have applied to interpretation of texts, cited in courts or disputes, that is, which were already issued. Again there is no mention of a requirement of general publication.
22That does not mean that edicts were not published: there was a custom of hanging these out on a public place (proponere) and the custom was kept, as we learn from the orders of the emperor in the posttheodosian Novels. The interest in the actual promulgation of new edicts in provincial towns, of which Matthews writes27, will have been real, people will certainly have been interested in reading it, it may also for a great part have been caused by the reverence for the emperor in any form which everybody was wise to show. Likewise curiosity will have waned soon too28, but all that does not answer the question. In general it will have been as today: many laws are unknown to people, but they go, if they are sensible, in case of need to a professional who knows the law and keeps up with it. The statement of the rabbi stands alone against the evidence, and the latter weigh far more.
23This conclusions also bear on the question, which text the compilers were to use. They were to gather the general rules issued. From the above it is clear that the letter of the emperor, once signed, was the normative text which should be used. Thus it sufficed for the compilers to peruse the copy books of out-going letters in the imperial archives, as was done in the case of the posttheodosian Novels over the years 438-448, which collection was amassed and sent to the west in 44829. But let us assume for one moment that if the actual provincial edict had been normative: it was this text the compilers were to use then. Would all these edicts have contained the same text, so that one provincial archive would have sufficed (apart from the problem, that some rule might have had a regional general application, so that the perusal of more than one archive would have been necessary)? If the governors had the authority to frame the actual rule in their own words, various formulations would be the result. It is, however, inconceivable that a governor would alter the imperial text. He would merely reproduce it and thus in every province the same text would be published30. Theoretically variations might occur, but these would be accidents or mistakes. Yet, that being so, there would have been no need to search in provincial archives for the final texts. A visit to the imperial archives, where the records of outgoing letters was kept, would suffice. Thus also in this case the outcome would be, that the compilers would use by far preference the central imperial archives.
24In this context Matthews cites post consulates as proof of a provincial origin. Laws issued at the end of the year would have arrived at their provincial destination early in the following year, when the new consul or consuls would not yet be known31. The latter is correct, as Bagnall et al. have shown32. Thus Matthews changes a post consular acceptance date wit Iul. into Ian., since the new consuls were normally known by July33. But the apparent assumption here is, that the acceptance date was written on the actual letter and that the letter was kept in the archive. However, it is not impossible that the letter was glued onto a roll, which had at the beginning a post consulate, and that in this way, at the moment of extraction, that post consulate was applied to the extract34.
25In the above we mainly dealt with the texts, contained in letters to high officials. They occupy most of the Theodosian Code. Other texts derive from imperial edicts, protocols etc. In such cases the compilers could also have contented themselves with consulting the imperial archives. Of course, if the imperial archives were incomplete, one had to look into other archives35. But that would have been the exception and not, as Seeck and now Matthews maintain, the regular case36.
Bibliographie
Des DOI sont automatiquement ajoutés aux références bibliographiques par Bilbo, l’outil d’annotation bibliographique d’OpenEdition. Ces références bibliographiques peuvent être téléchargées dans les formats APA, Chicago et MLA.
Format
- APA
- Chicago
- MLA
Bagnall et al. 1987
R. S. Bagnall, A. D. E. Cameron, S. R. Schwartz, K. A. Worp, Consuls of the Later Roman Empire, Atlanta, 1987 (Philological Monographs of the APhA, 36).
Kaiser 2010
W. Kaiser, «Zum Zitpukt des Inkrafttretens von Kaisergesetzen unter Justinian», ZSS-RA, 127, 2010, p. 172-201.
Kaser 1971
M. Kaser, Das römische Privatrecht, I. Das altrömische, das vorklassische und Klassische Recht, München, 1971.
Kaser 1975
M. Kaser, Das römische Privatrecht, II. Die nachklassischen Entwicklungen, München, 1975.
Lenel 1927
O. Lenel, Das Edictum Perpetuum, Leipzig, 1927 (3. Auflage).
Matthews 2000
J. Matthews, Laying Down the Law. A study of the Theodosian Code, New Haven & London, 2000.
Seeck 1919
O. Seeck, Regesten der Kaiser und Päpste für die Jahre 311 bis 476 n. Chr. Vorarbeit zu einer Prosopographie der christlichen Kaiserzeit, Stuttgart, 1919.
Sirks 1993
A. J. B. Sirks, «The Sources of the Code», dans J. D. Harries, I. N. Wood (éd.), The Theodosian Code. Studies in the Imperial Law of Late Antiquity, London, 1993, p. 45-67.
Sirks 2001
A. J. B. Sirks, «Making a Request to the Emperor: Rescripts in the Roman Empire», dans L. de Blois (éd.), Administration, Prosopography and Appointment Policies in the Roman Empire, Proceedings of the first workshop of the international network Impact of Empire, Amsterdam, 2001, p. 121-135.
Sirks 2003
A. J. B. Sirks, «Observations on the Theodosian Code: lex generalis, validity of laws», dans G. Crifò, S. Giglio (éd.), AARC, XIV, Napoli, 2003, p. 155-163.
Notes de bas de page
1 See for this figure Sirks 1993, p. 52.
2 Sirks 1993, p. 48 provides the theoretical structure of the texts found by the compilers, as assumed by Seeck, and mentions the errors the compilers could commit according to Seeck. But if we accept his idea of the way the compilers worked, we can never be sure that the date of a data is the original date of that data.
3 Matthews 2000, p. 289, note 27, referring also to Honoré, who assumed a shift in favour of eastern archives after 398.
4 Seeck 1919, p. 2, l. 12; 16-17; 24; 35.
5 Matthews 2000, p. 12-13.
6 See Sirks 2003.
7 Matthews 2000, p. 13-14.
8 See Sirks 2001.
9 Matthews 2000, p. 16-17.
10 Papirius Iustus, De constitutionibus; Paulus, Sententiarum libri V.
11 Dig IX, 1, 1, pr. The actio legis Aquiliae must have contained the basic elements of the lex Aquilia, or references to it, since here (in Dig IX, 2) the actual act is cited, thus giving the impression that the act itself was central. Yet everything depends on the application by the praetor. Consequently Lenel 1927 assumed that the actio for the first chapter of the law contained the words iniuria occidisse, the actio for the third chapter the words damnum corpori datum, which would explain the straight references to the law itself. Likewise the actiones utiles necessitated this recourse. Yet if we read the text itself, we see it might have contained the very demonstratio of the actio with si quis alteri... damnas esto (Dig IX, 2, 27, 5). In any case it is clear that only by means of an actio the law would have any effect.
12 Not in order to obtain validity, but as a custom.
13 CJ 6, 23, 7 (a. 290); 9 (a. 531).
14 See Sirks 2003.
15 CTh I, 4, 3.
16 CTh I, 1, 5.
17 CJ 1, 26, 2. If they were generalis and not contrary to laws or constitutions, and not superseded by posterior imperial authority, they had to be obeyed.
18 CTh I, 1, 6.
19 Matthews 2000, p. 66.
20 Matthews 2000, p. 169-170. He cites p. 169, note 2 my scheme of the «full panoply of annotation», but I did not state that this was the annotation, merely that it so was according to Seeck.
21 Matthews 2000, p. 187.
22 Matthews 2000, p. 187.
23 Kaser 1971, p. 242, note 60 and 61; Kaser 1975, p. 55, note 22 and 23.
24 Prescriptions like: this act will enter into force x days after its publication in the official journal.
25 Seeck as cited in Sirks 1993, p. 48, note 25. It is a false assumption, first because publication was not a prerequisite for validity (this is a modernism of Seeck) and by doing this the compilers, if such a publication would have been a prerequisite, would effectively have abolished it (which is highly unlikely), secondly because it would be a problem anybody could have conceived of before the designing of the Code and it would have been dealt with in CTh I, 1, 5, thirdly because in that case the best source would have been the emperor’s copy book of outgoing letters where the dates of issue were also registered.
26 One might wonder at what exact moment the law took effect, if it were not recited anymore by the lawgiver, as in the case of speeches before the senate, which usually were held by a prefect. If not all letters were subscribed by the emperor, who, as we know from the posttheodosian Novels, might add greetings like Deus te seruet per multos annos, and this probably was a mere sign of esteem, then the signing had no significance. We must assume then, that the order to send the text away, or the moment the text was approved and routine would set in, was the moment.
27 Matthews 2000, p. 187-189.
28 Matthews 2000, p. 190.
29 See Sirks 1993, p. 54, note 62 and p. 55.
30 Likewise it would not matter for the actual text whether the posted letter or the outgoing letter was used: of both the contents as regards the rule would be identical.
31 Matthews 2000, p. 212-213 and 268-269.
32 Bagnall et al. 1987.
33 Matthews 2000, p. 269.
34 See Sirks 1993, p. 54, note 62 and p. 55 for the similar way in which the et qui fuerit nuntiatus formula got into the posttheodosian novels from imperial copybooks.
35 That might explain the ca. 10% constitutions of obvious provincial extraction.
36 After finishing this contribution an article by W. Kaiser has appeared, in which he submits that under Justinian the actual publication was required for validity of a law, and that this indeed implied different provincial moments of gaining validity (Kaiser 2010). I hope to return to this later on.
Auteur
University of Oxford, Fellow of All Souls College
Le texte seul est utilisable sous licence Licence OpenEdition Books. Les autres éléments (illustrations, fichiers annexes importés) sont « Tous droits réservés », sauf mention contraire.
Les entrepreneurs du coton
Innovation et développement économique (France du Nord, 1700-1830)
Mohamed Kasdi
2014
Les Houillères entre l’État, le marché et la société
Les territoires de la résilience (xviiie - xxie siècles)
Sylvie Aprile, Matthieu de Oliveira, Béatrice Touchelay et al. (dir.)
2015
Les Écoles dans la guerre
Acteurs et institutions éducatives dans les tourmentes guerrières (xviie-xxe siècles)
Jean-François Condette (dir.)
2014
Europe de papier
Projets européens au xixe siècle
Sylvie Aprile, Cristina Cassina, Philippe Darriulat et al. (dir.)
2015