right and ought to be done. Now the peoples of whom the empire was composed were so diverse that they had but one characteristic in common, viz. that they were not Romans. They could have no common legal habit as a nation, and thus could generate no new positive customary law save merely local rules; the only influence they could unite to exercise was a negative one, to abandon all such portions of the system as were purely and exclusively Roman. And this destructive work, as we have seen, they did do. So with the other sources of the ius scriptum which are enumerated in the Institutes (i. 2. 3). Leges proper, plebiscita, and senatus consulta, as we know, had long been obsolete; and by the ordinance cited a few lines above, Constantine had given the last blow to the edicta magistratuum. Still, the development of law might have been carried on, as it had been from Augustus to Maximian, by the responsa prudentium and systematic treaties of authorized jurists; but we look in vain for these forms of legal activity, for true jurisprudence had already died a natural death. The juristic literature of the period now under consideration consists merely of compilations from statutes or constitutions and from the writings of earlier prudentes. Theodosius II (A.D. 408-450) attests the cessation of true juristic work in the enactment by which he promulgated his codex constitutionum : 'saepe nostra clementia dubitavit, quae causa faceret, ut tantis propositis praemiis, quibus artes et studia nutriuntur, tam pauci raroque extiterint, qui plena iuris civilis scientia ditarentur, et in tanto lucubrationum tristi pallore vix unus aut alter receperit soliditatem perfectae doctrinae.' His own explanation of the phenomenon is the vast multiplicity of imperial ordinances and juristic books without an adequate study of which a knowledge of the law was impossible; and the hope is expressed that a remedy for this will be afforded by the comparatively small dimensions of his own compilation.

Direct imperial legislation thus remained the only form in which changes could be effected in positive law. What Justinian says in Cod. 1. 14. 12 is true of the whole period extending backward from himself to Constantine: 'In praesenti leges condere soli imperatori concessum est, et leges interpretari solum dignum imperio esse oportet;' and this will account for the ever-increasing activity of the imperial legislation. In the interval of 130 years between Constantine and Theodosius II the mass of general constitutions had already become so great that the latter emperor's compilation formed a book of very considerable dimensions, and yet in this much was curtailed, still more omitted; for instance, it contained none of the enactments

of Diocletian, whose constitutions in the code of Justinian number

1247.

We have already touched upon the practical difficulties which arose in the application of responsa and juristic writings to the judicial decision of actions which came before the courts, and noticed the usage which prevailed as to this matter in the earlier empire, and which was directly confirmed by a rescript of Hadrian (p. 52 supr.). These difficulties had so increased with the growth of authoritative legal literature that it became imperatively necessary to discover some remedy. That which most immediately suggested itself was to limit the authority which this literature in general enjoyed to a moderate number of books, a plan which had already in its favour the rough scheme of precedence which general practice had established among the works of the classical jurists. Certain books had acquired a prescriptive authority in the various schools of law, public and private, which were located in various parts of the empire. A student commenced with the Institutes of Gaius, which were followed by the Commentaries on the Edict, especially those of Ulpian and Paulus, and the latter's responsa, quaestiones, and receptae sententiae: the course closed with Papinian and Modestinus. The writings of these five jurists had thus obtained a preponderance which naturally extended itself from the schools to the tribunals. Yet among them there were a great many controversies, and in this epoch neither the emperor nor the parties to an action were likely to be satisfied with the rule which Gaius (i. 7) states on the authority of Hadrian, that where jurists of different views were cited the judge might adopt whichever of the conflicting views he pleased. In order to obviate these difficulties, Constantine had (A.D. 321) deprived of all validity the notes of Ulpian and Paulus on Papinian, in which they had often found occasion to controvert his opinions. A more complete remedy was provided by the so-called 'law of Citations,' issued A.D. 426 by the emperors Theodosius II and Valentinian III. By this it was enacted that when conflicting opinions were cited in a court of justice from juristic writings with statutory authority, that opinion should be adopted by the judge in whose favour most jurists were brought forward; if the numbers were equal, the view of Papinian should prevail; if, in case of equality, no definite opinion on the case in hand could be extracted from this jurist, the judge might select from the conflicting views at his own discretion. It thus became the business of the parties to relieve the judge of the labour of collating the 'iuris auctores.' The procedure is compared by Rudorff to that of the

imperial council, or court of appeal, in which the majority of votes had always won the day; 'when it had ceased to be usual to convoke and consult living jurists, the practice was transferred to the councils of the dead.' So far the law of Citations' creates no difficulty; but as to the effect of another passage there has been much diversity of opinion. That passage runs as follows: Papiniani, Pauli, Gaii, Ulpiani atque Modestini scripta universa firmamus, ita ut Gaium, quae Paulum, Ulpianum et cunctos comitetur auctoritas, lectionesque ex omni eius opere recitentur. Eorum quoque scientiam, quorum tractatus atque sententiae praedicti omnes suis operibus miscuerunt, ratam esse censemus, ut Scaevolae, Sabini, Iuliani atque Marcelli, omniumque quos illi celebrarunt, si tamen eorum libri, propter antiquitatis incertum, codicum collatione firmentur. Ubi autem diversae sententiae proferuntur,' etc. (Cod. Theod. 1. 4. 3). This is explained by some writers1 to mean (1) that statutory force was hereby for the first time given to the works of Gaius, Papinian, Ulpian, Paulus, and Modestinus, and to such opinions of other prudentes as were therein cited; and (2) that all other juristic literature (including the notes of Ulpian and Paulus on Papinian) was in future to have no validity whatever. If this interpretation is the true one, the view which has been above stated, that the writings of authorized jurists had always the statutory force enjoyed by their responsa, is clearly erroneous. But the arguments against it are extremely strong. Though the 'law of Citations' confirms the authority of the five leading jurists, it is very far from expressly excluding the authority of others; it mentions others, and clearly only exempli gratia, as representative of the great mass of other juristic literature; and the weight which it allows to these others is not less, except in cases of equality of voices, than that of the five. Secondly, it is to be observed that the works of other jurists are frequently cited as authorities by later emperors (e. g. Julian by Leo and Anthemius in Cod. 6. 61. 5; Marcian by Justinian in Cod. 7. 7. 1; Tertullian in Cod. 5. 70. 7). But the strongest argument of all is to be found in the Constitutions issued by Justinian with reference to the compilation of the Digest, in which he instructs the compilers to make excerpts from the authorized jurists, and in the fact that, acting under this commission, they selected thirty-nine writers, and inserted in the work passages from every one of them; this they could not possibly have done if the only autho

1 E. g. Savigny, System i. § 26; History of the Roman Law in the Middle Ages, p. 28; Marezoll, Institutionen § 29.

rized jurists at the time of their appointment had been the five whose names have been mentioned. And the force of this fact is irresistibly increased by the words used by Justinian in Cod. 1. 17. 2. 20 with reference to the publication of the Digest, where he says that he has therein made selections from the jurists (legislatores et commentatores) 'quos et anteriores piissimi principes admittere non sunt indignati.'

It would seem then that the true meaning of the disputed passage is that which has been given to it by other civilians'; namely, that it was intended to supply a criterion by which the authority of any and every juristic writer could be tested. Its intention was not to exclude any authorized jurist, but to remove the difficulty which arose from the existence and quotation of works by writers who had received no auctoritas or ius respondendi, and to which Justinian bears witness in Cod. 1. 17. 1. 4. The test was to be the ius respondendi, and the possession of this was to be inferred from the fact of one jurist being cited as an authority by another of undeniable reputation. The legislator attained his object by selecting a small number of the most recent and distinguished jurists who had possessed the ius respondendi, and by confirming their authority, and also that of all others whose views they quoted. Some difficulty is occasioned by the words' si tamen eorum libri propter antiquitatis incertum codicum collatione,' which seem to mean that even when a jurist is cited by one of the five, he is not to be allowed a voice in the decision of a case unless his view, as quoted, is proved and confirmed by a comparison of the passage of the citing jurist with the original work (codices) of the author quoted. The enactment would thus practically secure to all approved writers the weight which they had hitherto enjoyed, while it would relieve judges of the weary task of listening to the opinions of others of only second-rate importance; it excluded only jurists who had never possessed a first-rate reputation, and in particular those who lived after Modestinus 2.

The only subject with which we have yet to deal, before proceeding

E. g. Puchta, Institutionen vol. i. p. 372, (8th edition 1875).

The opening words in the passage transcribed above from the 'law of Citations,' by which the statutory authority which the other four leading jurists had always possessed is extended to Gaius, go to show that the latter was not a 'iuris auctor,' and had never been invested with the 'ius respondendi.' This inference is supported by the fact that though his writings were so various, they contained no responsa, though among them was a 'liber de casibus,' in which he collected a large number of illustrative cases, many of which were imaginary, while of the real ones none had been authoritatively pronounced upon by himself.

to speak of the legislation of Justinian, is that of the codices constitutionum, or compilations of imperial ordinances, which spread themselves over about two and a-half centuries. The first with which we are acquainted is the codex Gregorianus, a collection of constitutions issued from the time, probably, of Hadrian up to A.D. 284, and principally rescripts of Septimius Severus and the emperors who succeeded him. Its exact date is uncertain, but it cannot well be placed earlier than A. D. 295; its arrangement was based upon that of the current commentaries on the Edict. Another compilation of the same character is the codex Hermogenianus, which may be regarded as an appendix to the foregoing. The oldest constitution which it contained was issued A.D. 290 or 291; the great majority of them belong to Diocletian and Maximian (A. D. 284-305), though seven enacted by Valens and Valentinian, about A. D. 365, are also referred to it; if this is correct, its composition must lie between A.D. 365 and a.d. 398, but there are civilians who place it many years before the earlier of these two dates, and accordingly deny that it could have comprised constitutions of Valens and Valentinian. Both of these codices, so far as we know, were formed by private persons, that is to say, not under the authority of an imperial commission; they were in constant use both in the West and in the East, and lost their importance in the latter half of the empire only by being absorbed in his own code by Justinian.

The contents of the so-called 'Vatican fragments' are more miscellaneous. They are portions of a compilation of excerpts from juristic writings and imperial rescripts and edicts, and were first published in 1823. The author is unknown, but the date of the work must be between A.D. 372 and A.D. 438, for it contains a constitution issued in the first of these two years, and in the latter the Theodosian code was published. The jurists whose works are most largely laid under contribution are Papinian (responsa and quaestiones), Ulpian (libri ad edictum, etc.) and Paulus (libri ad edictum, sententiae, responsa, quaestiones, etc.); Celsus, Julian, and Pomponius are also often cited. The work seems to have had a purely practical purpose, and was arranged in Titles subdivided into rubrics. Seven of the Titles are extant, relating to the following subjects: (1) ex empto et vendito; (2) de usufructu; (3) de re uxoria et dotibus; (4) de excusatione; (5) quando donator intelligatur revocasse voluntatem ; (6) de donationibus ad legem Cinciam; (7) de cognitoribus et procuratoribus.

In A. D. 429 Theodosius II, Emperor of the East, informed the