CONJECTURES AND CRITICISM IN BOOK 1 OF THE CODEX JUSTINIANUS

Since 2007, a team of American and British ancient historians has been preparing a new translation of the Codex Justinianus. The ‘Codex Project’ was launched by chief editor Bruce W. Frier; the goal of the project is to create the first reliable English translation of the Codex Justinianus on the basis of the standard edition by Paul Krüger.² Since 1932, the notoriously unreliable translation by Scott has remained the only one in English.³ The new translation by the Codex Project should appear soon.⁴

The translation of the Codex Project will not, however, be created ex nihilo. The work has been expedited by the existence of an unpublished translation by Justice Fred H. Blume, whose work has served as the basis of the text. The remarkable story of the young immigrant from Winzlar, Germany, is well worth reading.⁵ Settled in Cheyenne, Blume served for 42 years on the Wyoming Supreme Court; in his spare time, from 1920 to 1952, he laboured over a commented translation of both the Codex Justinianus and the Novels of Justinian. Blume also collaborated with Clyde Pharr on the English translation of the Codex Theodosianus, which remains the essential translation today.⁶ But while Pharr saw the translation of the Codex Theodosianus into print, Blume's translation of the Codex Justinianus remained unpublished: Pharr had planned a series of English translations of legal texts, but only his own Theodosian Code and Ancient Roman Statutes by Johnson, Coleman-Norton and Bourne appeared.⁷

That will soon change. In 1971, Blume left his personal library to the University of Wyoming; more than 30 years later, Timothy Kearley, Director of the Law Library, rediscovered Blume's typewritten translation, digitized and edited the 4,521-page manuscript, and has made both versions available online.⁸ On the basis of Blume's complete translation, and with the blessing of Blume's estate (his nephew, Frederick Paul Blume), the Codex Project began to review and revise the translation and at last prepare it for publication.

As the editor of Book 1 of the Codex Justinianus, I have had ample opportunity to scrutinize the Latin and Greek text on which the translation is based. In the following, I present my reflections on a number of passages in which I have introduced emendations or editorial changes to the standard text of the Codex Justinianus established by Paul Krüger. Obviously, since I depend on Krüger's apparatus criticus for the readings of the most important manuscripts and have not consulted them myself, it will remain for a future editor ultimately to decide how convincing the following proposals are. The needs of a translator, however, are different from those of an editor. In several passages, I have chosen alternatives for reasons of intelligibility and coherence, where a conservative editor might let Krüger's text stand. Until a new editor, however, replaces Krüger's now 135-year-old major edition and re-evaluates the evidence of the manuscripts and other witnesses,⁹ it must suffice to explain here the minor changes I have made to the text for the purpose of preparing the English translation of Book 1.

The passages discussed below are: CJ 1.2.15, 1.3.32.7 (and 1.2.17.2), 1.3.45.1, 1.3.53.1 (and 9.13.1.1), 1.4.34.12, 1.12.6.9, 1.17.2.9, 1.17.2.10, 1.17.2.16, 1.23.7 pr., 1.26.2, 1.27.2.1, 1.27.2.18, 1.34.2, 1.49.1.2. A summary of the passages and the proposed changes appears in the form of a table at the end of this article.

1. CJ 1.2.15 (22.33 K.)

CJ 1.2.15 begins as follows in Krüger's edition:

[Αὐτοκράτωρ Ζήνων Α.] ... Εἴ τις δωρεὰν κινητῶν ἢ ἀκινήτων ἢ αὐτοκινήτων πραγμάτων ἢ οἱουδήποτε δικαίου ... τὴν αὐτὴν δωρεάν ... κρατεῖν καὶ ἐν ἀπαιτήσει ἐκ παντὸς εἶναι τρόπου ...

[Emperor Zeno Augustus] ... If anyone should donate immovable, movable, self-moving, or any category of property ... the donation ... (shall be) valid and subject to collection ...

Since the Greek constitutions of the Codex Justinianus were lost in the Latin manuscript tradition, Krüger restored what remained of them from a variety of later Greek collections. He restored CJ 1.2.15 from the Nomocanon XIV Titulorum, a synthesis of secular and ecclesiastical law assembled in the early seventh century.¹⁰ The inscriptio of the constitution, which indicates the emperor who promulgated it, is missing in all witnesses, including the Nomocanon, but Krüger inferred from the following constitution that Zeno was the author and restored his name accordingly.¹¹ One further supplement may be made, though. In the process of excerpting and potentially adapting this passage, the editor of the Nomocanon omitted a subordinating verb that would account for the infinitives κρατεῖν and εἶναι in the surviving text. A literal translation of these words would give ‘be valid’ and ‘to be’, not ‘shall be’ as supplied above. For this reason, and to make the passage more intelligible, I also supply the verb Θεσπίζομεν (‘We decree’ = Latin sancimus) before the extant passage. A form of this or a similar verb must have stood somewhere in the original constitution, although not necessarily as the first word.¹²

Proposed text:

[Αὐτοκράτων Ζήνων Α.] ... [Θεσπίζομεν] εἴ τις δωρεὰν ...

[Emperor Zeno Augustus] ... [We decree that], if anyone should donate ...

2. CJ 1.3.32.7 (39.17 K.) AND CJ 1.2.17.2 (24.18 K.)

In this constitution, Leo and Anthemius (a. 472) confirm the privileges of various Christian churches and charitable houses, among them xenodochia and ptochia (n. pl.), that is (pilgrim) hospices and poorhouses.¹³ The superintendents of such institutions are freed from public liturgies (munera). Alongside bishops and their clerics or monks, the text lists the superintendents of hospices (xenodochos) and, surprisingly, beggars (ptochos):

Privilegiis sane singulis quibuscumque sacrosanctis ecclesiis orthodoxae fidei, xenodochiis sive ptochiis tam generaliter quam specialiter attributis perpetuo reservandis, nullis eas earumque sacerdotes aut clericos cuiuslibet gradus aut monachos vel ptochos aut xenodochos orthodoxae fidei deputatos extraordinariis muneribus praecipimus praegravari.

The individual privileges conferred on the various holy churches of the orthodox faith, hospices, or poorhouses, both generally and specifically, shall be maintained permanently. We order that neither they nor their bishops or clerics, of whatever rank, or monks, or beggars or the superintendents of hospices, of orthodox faith, should be burdened by assignment to extraordinary liturgies.

The word in bold, ptochos – that is, the Greek word πτωχούς – must be a mistake; it is otherwise unattested in Latin.¹⁴ Krüger indicates in the apparatus that codices C and originally also R had the reading ptochios, whereas codex M read ptochis, before it too was emended to read ptochos.¹⁵ Most probably, a mistake of haplography very early in the manuscript tradition caused the normal word for the superintendent of a poorhouse in the accusative plural, ptochotrophos,¹⁶ to be shortened inadvertently to ptochos. A scribe probably leapt from the h in the ch of ptocho- directly to the end of the word, the -hos of -trophos, without realizing he had omitted the intervening letters.¹⁷ However the reading came about, the passage must be emended one way or another, since the confusion of the superintendent of a charitable institution with its beneficiaries is inexplicable, and it makes no sense at all to protect beggars against extraordinary liturgies they had no means of fulfilling.¹⁸ The Greek translation in the Nomocanon (9.1, p. 534 Pitra), moreover, reads τοὺς ἐν τοῖς πτωχείοις ἢ ξενοδοχείοις ἀφωρισμένους (‘those assigned to hospices or poorhouses’), subsuming the superintendents of both institutions in one construction. Hence a Latin original in which ‘beggars’ or ‘the poor’ took the place of the superintendents of poorhouses alongside the superintendents of hospices is unlikely.

Proposed text:

... aut monachos vel ptochotrophos aut xenodochos orthodoxae fidei ...

... or monks, or superintendents of poorhouses or hospices, of orthodox faith ...

CJ 1.2.17.2, from an undated constitution of Anastasius (491–518), presents a similar case, in which the superintendent of a πτωχεῖον, here called the διοικητής, is ordered to be present at the authentication of the endowment of the poorhouse together with his ministers ‘and the poor’ (καὶ τῶν πτωχῶν).¹⁹ It seems better to read πτωχοτρόφων here too than to assume that the names of beggars really were registered alongside those of the superintendent of the poorhouse and his staff. If the poor were required to attend, one might expect in the same passage by analogy that the guests at the hospice or even the orphans at the orphanage would similarly be summoned; obviously, that is not the case. The explanation by haplography offered above, however, does not apply to the Greek version of the constitution, §2 of which Krüger restored from a later sixth-century collection, the Collectio Tripartita. There is no recorded variant for the passage in this version.²⁰ The manuscript reading would be correct, then, with respect to the Collectio Tripartita, but it seems fair to suspect that a mistake identical to that proposed for CJ 1.3.32.7 had already been made in the lost Latin original of CJ 1.2.17.2.²¹

3. CJ 1.3.45.1 (48.11 K.)

CJ 1.3.45.1 has been restored from both the Nomocanon and the Collectio XXV Capitulorum.²²

Ἀλλ’ εἰ μὲν οἰκοδομὴν ἐκκλησίας ἐπιτάξειεν ὁ διαθέμενος, εἴσω τριετίας πάντως παρασκευάζειν αὐτὴν πληροῦσθαι, εἰ δὲ ξενῶνος ποίησιν, εἴσω ἑνὸς ἐνιαυτοῦ μόνον τοῦτο ποιεῖν ἀναγκάζεσθαι, ὡς ἱκανοῦ τοῦ χρόνου τούτου καθεστῶτος πληρῶσαι τὰ βεβουλευμένα τοῖς τετελευτηκόσι ...

If the testator orders the construction of a church, they (i.e. the heirs) shall see to it that it is completed entirely within three years; if the building of a hospice, they shall be compelled to do this only within one year, for this is a sufficient period of time for carrying out the wishes of the deceased ...

In this passage, if the deceased instructs his heir to build a church, the heir has three years to build it; if the deceased wants a hospice, does the heir have only one year to build it, or does he have a year to do only that (that is, build a hospice exclusively)? As the text presently stands (μόνον τοῦτο), it suggests that the heir may do nothing in the year concerned other than build the hospice. Emending μόνον, however, by putting it in the genitive case μόνου in agreement with ἑνὸς ἐνιαυτοῦ would make much more sense, and the combination of these terms is attested elsewhere in Justinian's Novels.²³ Μόνου moreover is the reading of the Collectio XXV Capitulorum.²⁴ The explanation that immediately follows in the text – that this amount of time (one year) should be enough to build the hospice – would have no purpose if the specific activity of the heir, not the time in which he completes it, was relevant. The point of the passage is not to limit the activity of the heir, but rather to fix the amount of time within which he must complete the project entrusted to him by the deceased. Unless the word μόνον should be removed entirely,²⁵ we might hypothesize that an early scribe changed the ending –ου to –ον to make it agree with τοῦτο.²⁶

Proposed text:

... εἴσω ἑνὸς ἐνιαυτοῦ μόνου τοῦτο ποιεῖν ἀναγκάζεσθαι, ὡς ἱκανοῦ τοῦ χρόνου τούτου καθεστῶτος πληρῶσαι τὰ βεβουλευμένα τοῖς τετελευτηκόσι·

... they shall be compelled to do this within one year alone, for this is a sufficient period of time for carrying out the wishes of the deceased ...

4. CJ 1.3.53.1 (55.15 K.) (AND CJ 9.13.1.1 [832.5 K.])

Punctuation is also interpretation. The meaning of this passage depends in part on the placement of a comma, and one widespread interpretation of it seems incorrect. I have loosely adapted Blume's original translation of CJ 1.3.53.1 and CJ 9.13.1.1 (which differ somewhat) to highlight the interpretation in question:

Qui itaque huiusmodi crimen commiserint et qui eis auxilium tempore invasionis praebuerint, ubi inventi fuerint in ipsa rapina et adhuc flagrante crimine comprehensi a parentibus sanctimonialium virginum vel viduarum vel diaconissarum aut earum consanguineis vel tutoribus seu curatoribus, convicti interficiantur.

Those who commit a crime of this kind and those who assist them at the time of the assault, if they are discovered in the act of ravishing and caught in flagrante, may, as convicted criminals, be killed by the parents of the holy virgins or widows or deaconesses, or by their relatives, tutores, or curatores.

As Blume translates, this constitution at first glance appears to grant the family and guardians of a ravished woman the right to kill raptores caught in the act, just as the paterfamilias or husband of a woman might kill an adulterer under the lex Iulia de adulteriis.²⁷ The resemblance, however, is misleading. The right to kill adulterers (ius occidendi) could be exercised under the lex Iulia only by the men closest to the women affected: either her father or her husband, and under conditions that tended to make its use unpleasant or hazardous.²⁸ Here, a much greater number of people are involved: the woman's parents and relatives (to what degree?), and even her legal guardians; in the parallel law CJ 9.13.1, which additionally concerns freedwomen and slaves, the list of potential killers includes even patrons and slave-owners (§1).

Whatever these persons were expected to do, however, must derive not from the verb interficiantur but from one of the preceding participles. Krüger's punctuation of this dense passage, with a comma before the word convicti, should lead the reader to connect the persons in the construction a parentibus ... consanguineis vel tutoribus seu curatoribus with the participle comprehensi. Accordingly, a raptor may be condemned and executed if he is caught (comprehensi) in the act by the parents, relatives or guardians of the woman. It makes little sense, though, why the perpetrator should be punishable only if he is detected and seized by the persons named. The punctuation of the passage is disputable also on stylistic grounds. The participial phrases in the clause ubi inventi fuerint in ipsa rapina et adhuc flagrante crimine comprehensi create an elegant chiasmus, while it is equally possible to connect the ablative of agent a parentibus, etc. not with the preceding participle comprehensi but with one of the passives that follow: the passive participle convicti or the passive verb interficiantur.

Although Blume understands convicti in a metaphorical sense, ‘as convicted criminals’,²⁹ this word should indicate that even raptores caught in flagrante should be tried.³⁰ The due trial of criminals caught in the act, including raptores, is attested in another constitution in the Codex Justinianus.³¹ Moreover, within the constitution, Justinian insists that raptores caught in the act but able to avoid capture should be condemned to death only ‘after legitimate proof recognized by the law’.³² The persons connected to the woman might have seized the raptor caught in the act, but, more importantly, they are expected to prosecute him, bringing about his conviction. In the case of the rape of a slave, for instance, it would necessarily have been the owner (dominus, CJ 9.13.1.1) who brought charges. In the longer, secular version of the law, Justinian similarly punishes parents who fail to prosecute the raptus of their daughter (CJ 9.13.1.3c), although he says nothing about her relatives or guardians. Primarily the parents or guardians must be responsible for the prosecution of the perpetrator. That the family or tutor/curator of a woman would undertake the prosecution in a criminal trial is attested by several examples.³³

It is thus all the more remarkable that the Greek version of this constitution in the Collectio Tripartita explicitly grants parents, siblings and guardians the right to kill not only the rapists caught in flagrante (ἐπ’ αὐτοφόρῳ), but also their accomplices.³⁴ The limited ius occidendi of the lex Iulia was subject to abuse:³⁵ to what extravagant bloodbaths could abuse of a carte blanche for multiple killings have led, if it really were law?

To return to the punctuation and translation of this passage: one could retain Krüger's punctuation and argue that the persons named were required to capture the offender; their involvement in the ensuing trial would be assumed but not stated. More probably, however, Justinian expects them to undertake the prosecution itself on behalf of the violated women. If so, then our comma should be moved after convicti and another inserted after comprehensi, creating a distinct appositional clause. Of course, if the Codex Justinianus was written in scriptura continua with punctuation only at the end of a fragment like the sixth-century Codex Florentinus of the Digest, ancient readers will have had to interpret the text for themselves.³⁶

Proposed text:

Qui itaque huiusmodi crimen commiserint et qui eis auxilium tempore invasionis praebuerint, ubi inventi fuerint in ipsa rapina et adhuc flagrante crimine comprehensi, a parentibus sanctimonialium virginum vel viduarum vel diaconissarum aut earum consanguineis vel tutoribus seu curatoribus convicti, interficiantur.

Those who commit a crime of this kind and those who assist them at the time of the assault, if they are discovered in the act of ravishing and caught in flagrante, (and/if) convicted by the parents of the holy virgins or widows or deaconesses, or by their relatives, tutores, or curatores, may be executed.

The same punctuation is also recommended in CJ 9.13.1.1, of which CJ 1.3.53 is a version adapted especially for consecrated virgins.³⁷

5. CJ 1.4.34.12 (73.5 K.)

This passage, from a constitution that appears both in the Nomocanon (9.27) and in the Collectio XXV Capitulorum (const. 21), concerns clergymen who are caught playing dice and engaged in other forms of gambling. The guilty are excommunicated or banished from the Church; §§11–13 threaten punishment for religious and secular authorities who fail to carry out Justinian's commands. What he commands is contained in §§9–10. A wealthy clergyman who proves to be an unrepentant gambler is to be enrolled in the curia of the city where he served as priest (τὸ τῆς πόλεως ἐκείνης βουλευτήριον, καθ’ ἣν ἱερᾶτο πρότερον, §10, p. 72, line 24), or even in the curia of another city that needs decurions, if his own city should be too small; a poor clergyman, on the other hand, is enrolled as a cohortalis on the staff of the governor (ταξεώτης τὸ λοιπὸν ἔσται τῆς ἐπιχωρίου τάξεως, ibid., lines 26–7). In both contingencies, the guilty party should be made to perform liturgies appropriate to his status. At the end of §12, however, where the punishments are briefly recapitulated, we read that one should see to it, ‘as has been established above’ (p. 73, line 4), that such a person is not delivered up to a curia or gubernatorial staff:

... καί, ἐπειδὰν παρ’ αὐτῶν ὁ ἐπὶ τοῖς τοιούτοις ἁλοὺς καθαιρεθείη, μὴ τῇ βουλῇ τῆς πόλεως ἢ τῇ τάξει τοῦτον παραδοθῆναι παρασκευάσαιεν.

... and when one convicted of such misconduct has been deposed, they shall see to it that he is not handed over either to the curia of the city or to the provincial staff.

The reading μή simply cannot be right. Krüger reports the reading ἤ in the Nomocanon, yet prefers the reading μή in the Collectio XXV Capitulorum.³⁸ In fact, the version of the constitution in the Nomocanon has neither ἤ nor μή in the corresponding place. There still is little reason to accept μή, however. The version of the constitution in the Nomocanon and Collectio Tripartita, though worded differently, clearly indicates that the offender must become either a curialis or a cohortalis.³⁹ The grammatical structure of the longer version also makes a parallel ἤ...ἤ construction likely. The reading ἤ is therefore recommended for μή retained by Krüger:

ἢ τῇ βουλῇ τῆς πόλεως ἢ τῇ τάξει τοῦτον παραδοθῆναι παρασκευάσαιεν.

they shall see to it that he is handed over either to the curia of the city or to the provincial staff.

Mommsen attempted to eliminate the inconsistency a different way: he proposed to add μή before the verb καθαιρεθείη.⁴⁰ This emendation would produce a parallel negative construction with μή and an optative verb: μὴ καθαιρεθείη and μὴ ... παρασκευάσαιεν. But it seems highly unlikely that someone caught in such vice (ὁ ἐπὶ τοῖς τοιούτοις ἁλούς) would not be deposed from office (μὴ καθαιρεθείη). The reading ἤ is clearly the better choice. Perhaps a scribe wrote μή for ἤ to avoid a presumed duplication of the eta in καθαιρεθείη.

6. CJ 1.12.6.9 (100.28 K.)

§9 of this constitution regulates how clergymen should act when unfree persons of various status seek refuge on holy ground. The scene is depicted very vividly. In Krüger's edition, we read:

Sane si servus aut colonus vel adscripticius, familiaris sive libertus et huiusmodi aliqua persona domestica vel condicioni subdita conquassatis rebus certis atque subtractis aut se ipsum furatus ad sacrosancta se contulerit loca, statim a religiosis oeconomis sive defensoribus, ubi primum hoc scire potuerint, per eos videlicet ad quos pertinent, ipsis praesentibus pro ecclesiastica disciplina et qualitate commissi aut ultione competenti aut intercessione humanissima procedente, remissione veniae et sacramenti interveniente securi ad locum statumque proprium revertantur ...

If, however, a slave, free or unfree tenant, servant or freedman, and any household member of this sort or bound to this status goes to sacrosanct places after breaking and pilfering certain things or having stolen himself, then immediately by the pious stewards or defenders, as soon as they are aware of the situation, namely through those to whom they belong, the fugitives may safely return to their proper place and condition after, in the presence of the stewards or defenders, according to ecclesiastical discipline and the nature of the crime, either appropriate punishment has been meted out, or most compassionate intercession has been made, and forgiveness has been granted by pardon and on oath …

On closer inspection, something is lacking. If slaves or coloni, etc. flee to a consecrated place, then the holy administrators or defenders of the church should – do what with them? No verb is provided to indicate what the fugitives may expect from the clergymen until we reach the word revertantur,⁴¹ where they are already on their way home. What happens before that point? It also remains unclear to what the clause per eos videlicet ad quos pertinent (‘namely through those to whom they belong’) refers. Mention of the persons responsible for the fugitives seems very abrupt. The sole main verb revertantur has to accommodate both the prepositional phrase a religiosis oeconomis sive defensoribus (‘by the pious stewards and defenders’) and the prepositional phrase per eos videlicet ad quos pertinent. This combination of modifying phrases does not make adequate sense. This messiness becomes particularly obvious when one tries to translate the passage: should the unfree persons return from the holy church administrators or defenders by the persons to whom they belong? Does the preposition a indicate movement away from the clergymen or, as appears at first glance, that something is done by the clergymen? The Latin in fact leads the reader to assume ‘by the holy church administrators or defenders’ until one arrives at the intransitive deponent verb revertantur.⁴²

This confusion is alleviated if one posits a small lacuna after the word potuerint in which a missing verb should be supplied. Probably, the fugitives should be reported by the clergymen, so that they may be retrieved by their owners or landlords. If one supplies the words nuntientur, ut, for example, the return of the fugitives is accomplished by the people who are responsible for them, while the clergymen (ipsis praesentibus) are in attendance to ensure that everything proceeds in an orderly fashion. The following paragraph of the constitution (§10), moreover, specifically stipulates that the clergymen should investigate fugitives and their cases and inform (instruant) the governors or ‘those persons to whom the cases and persons pertain’ (eos, ad quos causae et personae pertinent). CTh 9.45.5, from the year 432, prescribes nearly the same procedure when a slave flees to a church.⁴³ The following text, therefore, is tentatively proposed:

... statim a religiosis oeconomis sive defensoribus, ubi primum hoc scire potuerint, <nuntientur, ut> per eos videlicet ad quos pertinent ... securi ad locum statumque proprium revertantur ...

... then they shall immediately <be reported> by the pious stewards or defenders, as soon as they are aware of the situation, <so that> the fugitives may safely return to their proper place and condition, namely through those to whom they belong ...

As soon as the clergymen are aware of the fugitives, they should report them, so that their owners or landlords may take them back.⁴⁴

7. CJ 1.17.2.9 (113.1 K.)

Discussion of the next three passages is complicated by the fact that they enjoy two different manuscript traditions. CJ 1.17.2 is better known as c. Tanta, the introductory constitution that explains the conception of the Digest and enacts it as law. The text may thus be compared directly with the famous Florentinus manuscript of the Digest, the only witness to c. Tanta outside the Codex Justinianus.⁴⁵ Krüger took two of the following readings explicitly from the Florentinus.⁴⁶

In the first case, Krüger prefers the reading omne of the Florentinus to the reading omnes preserved in the Codex Justinianus. In the passage concerned, the members of the commission that prepared the Digest are named and praised:

… nec non Cratinum virum illustrem et comitem sacrarum largitionum et optimum antecessorem huius almae urbis constitutum: qui omnes ad praedictum opus electi sunt una cum Stephano, Mena, Prosdocio, Eutolmio, Timotheo, Leonide, Leontio, Platone, Iacobo, Constantino, Iohanne, viris prudentissimis; qui patroni quidem sunt causarum apud maximam sedem praefecturae, quae orientalibus praetoriis praesidet, omne autem suae virtutis testimonium undique accipientes et a nobis ad tanti operis consummationem electi sunt.

... and Cratinus, vir illustris, Count of the Imperial Finances and outstanding professor of This Generous City. All these were chosen for the above mentioned work together with Stephanus, Menas, Prosdocius, Eutolmius, Timotheus, Leonides, Leontius, Plato, Jacobus, Constantine, John, most learned men, who are advocates at the supreme tribunal of the prefecture that presides over the governors of the Orient; but they have gathered all testimony of their excellence from all quarters and were chosen by Us to carry out this work so great.

The most important witnesses for the word omne are divided: the Florentinus and the first reading of the second-most-important manuscript of the Codex (R) read omne; the Montecassino codex (C), on the other hand, the corrected reading of R, and the Montepessulanus codex (M) read omnes.⁴⁷ At first glance, it is not clear which variant is preferable. A parallel, however, may be identified in the preceding passage qui omnes ... electi sunt, as well as in the immediately following statement et, cum omnes in unum convenerunt gubernatione Triboniani. Justinian refers several times to all members of the Digest commission. The Greek version of the passage in the constitution Δέδωκεν appears to confirm this interpretation. Δέδωκεν reads:

… οἵπερ πάντες (= omnes) ῥήτορες μέν εἰσιν τῶν ἐνδοξοτάτων ὑπάρχων τῶν ἱερῶν ἡμῶν πραιτωρίων, δόξαν δὲ παρὰ πᾶσιν (= undique) τὴν ἐπὶ σοφίᾳ δικαίως ἔχοντες εἰχότως παρ’ ἡμῖν ἐκρίθησαν ἄξιοι καθεστάναι τοῦ τοσούτου μετασχεῖν ἀγῶνος.

… who all are advocates of the Prefects of Our most glorious sacred Praetoria and rightly hold esteem from all quarters, have appropriately all been judged by us to be worthy to partake in so great a task.

Although the placement of πάντες in Δέδωκεν would lead one to expect omnes before patroni in the Latin version, its later placement appears to be caused by the rephrasing of the Latin version as an antithesis with two verbs (sunt and electi sunt); qui gives the subject, while omnes in agreement with it allows natural use of the adversative, postpositive participle autem (instead of an awkward et qui). In Greek, πάντες stands next to the first of two participles that precede the main verb. The phrase παρὰ πᾶσιν in Δέδωκεν, on the other hand, should not be taken as equivalent to omne: it is translated by undique.

Moreover, the combination omne ... testimonium is contradicted by what is said immediately after: if the men had already received all testimony (i.e. recognition; Δέδωκεν: δόξαν) of their excellence or ability, their inclusion in the commission to prepare the Digest would not be an additional distinction. It is obvious, though, that Justinian wants to have both: all the men have demonstrated their outstanding talent, and additionally they have been elected by Justinian to compile the Digest. The appointment ensues from their existing prestige yet also adds to it. The reading omnes is thus not only stylistically, but also logically, superior. We may therefore prefer the reading of the tradition of the Codex to that of the Florentinus.

Proposed text:

... omnes autem suae virtutis testimonium undique accipientes ...

... all have gathered testimony of their excellence from all quarters ...

8. CJ 1.17.2.10 (113.16 K.)

A single letter can make the difference between the perfect subjunctive fuerit and pluperfect indicative fuerat. In the following excerpt, Justinian describes what was removed and what was retained in the legal texts incorporated in the Digest. These texts include not only works of Roman jurisprudence, but also imperial constitutions that the jurists may have cited in their works. Justinian explains that even when an imperial constitution was cited in the old writings of the jurists, he did not spare its text, but emended it as necessary. He let the names of the emperors stand, however, and adds that

quidquid legum veritati decorum et necessarium fuerat, hoc nostris emendationibus servavimus.

We exempted from Our emendations whatsoever had been appropriate and indispensable for the authenticity of the laws.

The word fuerat places the content that was pertinent and necessary for the authenticity of the imperial constitutions in the past before Justinian's intervention. Some part of these texts had (once) been necessary to ensure their authenticity; it follows, however, that when Justinian set about to emend such constitutions, this content was no longer necessary. On the contrary, though, Justinian is not referring to relevance that had once existed, and exists no more, but to relevance that still exists as the text is edited. It is because such content remains important that Justinian retains it in the Codex. What is decorum et necessarium must be recognized as contemporary with the revision of the text in the immediate past, and it remains decorum et necessarium in the present. The same thought is expressed more densely in Δέδωκεν: there Justinian makes the ‘truth’ of the laws his own.⁴⁸ This chronological distinction is expressed in Latin by the variant fuerit.

Proposed text:

quidquid legum veritati decorum et necessarium fuerit, hoc nostris emendationibus servavimus.

We exempted from Our emendations whatsoever was appropriate and indispensable for the authenticity of the laws.

As far as the manuscripts are concerned, the better manuscripts of the Codex Justinianus (Krüger's C and R, Mommsen's S)⁴⁹ have the reading fuerit. In his introduction to the Digest, Mommsen rated their consensus as ‘non minoris auctoritatis quam est Florentina’.⁵⁰ In a footnote, he listed examples in which the readings of the Codex tradition are superior to those of the Florentinus.⁵¹ The reading fuerit in CJ 1.17.2.10 may be added to the list.

9. CJ 1.17.2.16 (114.34 K.)

In the last passage from c. Tanta, Justinian concedes that something that should have been included in the Digest might have been overlooked:

Sed et si quid forsitan praetermissum est, quod in tantis milibus quasi in profundo positum latitabat, et, cum idoneum fuerat poni, obscuritate involutum necessario derelictum est: quis hoc apprehendere recto animo possit?

But even if something has been overlooked, which lurks in the depths among so many thousands (of books) and, although it was suitable to include, was inevitably left covered in darkness — who in his right mind could understand (seize?) this?

According to this reading, Justinian asks how any reasonable person can understand (apprehendere; less likely: ‘seize upon’) such an omission.⁵² Both Krüger and Mommsen remark in the apparatus ‘reprehendere (dett.)’.⁵³ In other words, reprehendere is the reading in less important or less reliable manuscripts. Their scepticism of apprehendere is justified. The word apprehendere is strange here, because the sentence seems to anticipate a verb of blaming, so that the apology⁵⁴ immediately after the passage has a point. To illustrate the argument from Justinian's perspective: how could a reasonable man not understand or not appreciate that in so large a compilation as the Digest something might have been overlooked? What is interesting is that the reading apprehendere appears to be original, which is precisely why it is retained by Krüger and Mommsen. It appears uniformly both in the Florentinus and in the Codex tradition. It is possible that a non has fallen out of the text, which would give the translation, ‘Who in his right mind could not understand this?’⁵⁵ This would preserve the reading apprehendere and give appropriate sense, but I am hesitant to adopt it. The kind of understanding expressed by apprehendo is more the result of a process of understanding than the process itself, more a ‘realization’ or ‘appreciation’ (or indeed a ‘grasping’) than simply ‘understanding’.⁵⁶ The qualification recto animo seems better suited to express incredulity at the audacity of potential critics rather than the obtuseness of those who fail to appreciate that mistakes inevitably are made in a massive undertaking.

Unless one assumes an unlikely error by the drafter of the constitution, the most plausible explanation for this strange locution is that it entered the tradition as a mistake made in the very first copies.⁵⁷ As Krüger indicates, the codices deteriores read reprehendere; I incline to think they are right. For the translation, I propose to read:

… quis hoc reprehendere recto animo possit?

… who in his right mind could fault this?

10. CJ 1.23.7 PR. (123.20 K.)

This constitution exhibits confusion of quae and quaevis and the coordinating conjunctions vel ... vel:

Vniversa rescripta, sive in personam precantium sive ad quemlibet iudicem manaverint, quae vel adnotatio vel quaevis⁵⁸ pragmatica sanctio nominetur, sub ea condicione proferri praecipimus, si preces veritate nituntur, nec aliquem fructum precator oraculi percipiat impetrati, licet in iudicio adserat veritatem, nisi quaestio fidei precum imperiali beneficio monstretur inserta.

We order that all rescripts, whether they issue out to petitioners or to some judge, which is named either an annotation or any pragmatic sanction, shall be produced on this condition, that the petition is grounded in truth; nor shall a petitioner reap any benefit from the oracle⁵⁹ obtained, though he alleges the truth in court, unless an inquiry into the veracity of the petition is shown inserted in the imperial indulgence.

The legislator apparently intends to regulate all rescripts, including adnotationes and pragmaticae sanctiones.⁶⁰ The wording is none the less strange. He begins with universa rescripta (‘all rescripts’), but then specifies one rescript that is called either an adnotatio or ‘any’ (quaevis) pragmatica sanctio. If the emperor intended to regulate only universae adnotationes or pragmaticae sanctiones, the scope of the constitution would be reduced significantly, since adnotationes and pragmaticae sanctiones were only two kinds of rescripts among several.⁶¹ But the purpose of the relative clause is not to limit the scope of the constitution to adnotationes and pragmaticae sanctiones, but rather to subject those forms explicitly to the enactment. It is one of the peculiarities of the late antique legislative style that universal declarations are reinforced with particulars, so as to emphasize and clarify what has already been established. Thus here the two rough categories of rescripts first are summarized: both those that are issued to private petitioners (in personam precantium) and those that are dispatched to public officials (especially provincial governors: ad quemlibet iudicem) on inquiry. To these, adnotationes and pragmaticae sanctiones should be added, which are valid only in connection with a single case. The relative pronoun quae is not neuter plural referring to universa rescripta, but feminine singular in agreement with the nearest substantive, adnotatio; the singular verb nominetur similarly agrees in number with its neighbour, pragmatica sanctio. The draftsman apparently chose a relative clause to avoid repeating the sive construction that he had used in the first half of the sentence.

The reading quaevis is more difficult. Although Krüger decided in favour of one of the attested variants, he was sufficiently sceptical to mention in the apparatus that Mommsen had proposed to read quaeve adnotatio quaeve pragmatica. Mommsen's instinct to emend the text is understandable. The unevenness of the phrases quae vel and vel quaevis is itself suspicious. Numerous passages in the Codex Justinianus show how the terms adnotatio and pragmatica sanctio are often discussed specifically together or, as here, with rescripta in general.⁶² It is thus legitimate to expect the disturbed passage in CJ 1.23.7 to correspond roughly to passages in which rescripts are first discussed generally and then both adnotationes and pragmaticae sanctiones are specified. CJ 1.2.10 (= 11.4.2) is the most similar passage to compare: nec si caeleste contra proferatur oraculum, sive adnotatio sit sive divina pragmatica, whereby the ‘heavenly oracle’ mentioned is none other than the sacred words of the emperor, a private rescript. In all comparable passages, repetition of the conjunctions vel and sive or seu is frequent; quaeve, on the other hand, appears in the Codex Justinianus only in the combination qui quaeve (‘who, man or woman’),⁶³ and quaevis appears in the Codex Justinianus only in CJ 1.23.7 pr. Hence Mommsen's attempt to tie rescripta, adnotatio and pragmatica sanctio together as three parts by means of -ve seems improbable, although he succeeds in making something intelligible of the manuscript readings quae vel and vel quaevis (or rather quae ius or quamvis). A different solution would be to assume that quaevis (or quamvis or quae ius) results from duplication of the first words of the phrase quae vel. The confused manuscript tradition may have arisen as various attempts were made to integrate the superfluous words intelligibly into the sentence.⁶⁴

Proposed text:

Vniversa rescripta, sive in personam precantium sive ad quemlibet iudicem manaverint, quae vel adnotatio vel pragmatica sanctio nominetur, sub ea condicione proferri praecipimus ...

We order that all rescripts, whether they issue out to petitioners or to some judge, whether named either an annotation or a pragmatic sanction, shall be produced on this condition ...

11. CJ 1.26.2⁶⁵ (125.6 K.)

This brief constitution may be cited in full:

Idem A. Restituto. Formam a praefecto praetorio datam, et si (etsi?) generalis sit, minime legibus vel constitutionibus contrariam, si nihil postea ex auctoritate mea innovatum est, servari aequum est. D. id. Aug. Severo et Quintiano conss.

The same Augustus to Restitutus. It is reasonable that a ruling (forma) given by the Praetorian Prefect, and if (etsi: even if?) it is general, not at variance with the laws or constitutions, should be observed if no innovation has been made subsequently by My authority. Given August 13, in the consulship of Severus and Quintianus (235).

The reading et si first appears in the stereotype edition of the Codex Justinianus. In the editio maior, one still reads etsi. In both cases, though, the meaning remains mysterious. The forma of the praetorian prefect should be understood as a verdict or decision, probably in the form of a document.⁶⁶ Apparently, the constitution revolves around the question to what extent such a decision may be cited as a precedent. If the reading should really be et si, not etsi, then one might have expected a corresponding et si before the words nihil postea, creating a ‘both–and’ construction: ‘both if it is generalis, <and> if no innovation has been made subsequently by My authority’, the forma of the prefect remains valid. Alternatively, if one interprets the reading as etsi (‘even if’), a negation would be expected, since a decision that may be qualified with adjective generalis should be suited to serve as a precedent. In that case, one would expect etsi <non> generalis sit. One last solution, and probably the simplest, is offered by isolated witnesses that omit the first et:⁶⁷

Formam a praefecto praetorio datam, si generalis sit, minime legibus vel constitutionibus contrariam, si nihil postea ex auctoritate mea innovatum est, servari aequum est.

It is reasonable that a ruling (forma) given by the Praetorian Prefect, if it is general, not at variance with the laws or constitutions, should be observed if no innovation has been made subsequently by My authority.

Or should one conversely place et before the words si nihil, ‘If it is generalis ... and if nothing has been decreed ... ’ etc.?

It may remain an open question which of the alternatives – et si ... et si, etsi <non>, si ... si or si ... et si – should be given preference, but they all are better than the reading printed in the standard edition. The Greek version of the constitution in the Basilika, however, the massive late ninth-century compendium of law compiled under Leo VI the Wise, might support the reading si generalis sit.⁶⁸ The Greek version reads,⁶⁹

Κρατείτωσαν οἱ τῶν ἐπάρχων τύποι γενικοὶ ὄντες.

The general edicts of the Praetorian Prefects shall be valid.

The force of γενικοὶ ὄντες is essentially the same as si generalis sit. Only the Vatican codex among the chief witnesses agrees – but that may perhaps tip the scales.⁷⁰

Proposed text:

Formam a praefecto praetorio datam, si generalis sit ...

... a ruling (forma) given by the Praetorian Prefect, if it is general ...

12. CJ 1.27.2.1 (131.26 K.)

In a forthcoming article on the Würzburg fragment of the Codex Justinianus, Simon Corcoran proposes a very attractive emendation.⁷¹ We presently read,

et armatas militias et duces militum ordinare disponimus.

We have resolved to organize both the armed forces and leaders of the soldiers.

The armies in question are those that should protect the newly recovered provinces of Africa. The expression ‘leaders of the soldiers’ is surprisingly simplistic, more at home in an elementary Latin textbook than in Justinianic law. Corcoran is undoubtedly right to see here (commenting on fol. 1^rb) a mistake for duces limitum (‘frontier commanders’), whom Justinian immediately begins to discuss: ‘We therefore decree that the dux limitis of the province of Tripolitana shall have his seat presently in the city of Leptis Magna’ (§1a). Corcoran notes that all the manuscripts, including the Würzburg fragment, read militum. The most likely explanation of the error is simple transposition of the l and m.

Proposed text:

et armatas militias et duces limitum ordinare disponimus.

We have resolved to organize both the armed forces and frontier commanders.

13. CJ 1.27.2.18 (134.20 K.)

Among the affairs that the reconquest of Africa brought with it, the salaries of the new provincial administration needed to be regulated.⁷² The salary of the dux and his staff should be met by local taxes. The amounts for the various officials are given in an appended list (§§19–34). The most diverse expressions are dredged up to express what the list shows, establishes or contains with appropriate variatio: §17: ... quantum subter adnexa declarat notitia (‘the amount that the schedule appended below shows’); §17a: ... in subdita notitia taxatum est (‘has been defined in the appended schedule’); ibid.: ... quorum nomina in subiecta notitia continentur (‘whose names are listed in the appended schedule’). Finally we meet an unusual use of the verb detinet:

Ad haec iubemus, ut deo iuvante unusquisque dux seu eorum officia, secundum quod notitia subter adnexa detinet, emolumenta sua ex tributis Africanae provinciae ex kalendis Septembribus instantis felicissimae tertiae decimae indictionis percipiant.

In addition, We order that, with the help of God, each commander or his staff shall, according to the amount that the schedule attached below ‘detains’, receive their perquisites from the tribute of the province of Africa from September 1^st of the coming, most prosperous thirteenth indiction.

The officials should receive the salaries that are listed in the notitia. But can detinet mean approximately the same thing as continet? Unfortunately not, which the variant continet mentioned by Krüger in the apparatus also shows. One might write the usage off as a mistake by the author of the constitution.⁷³ There is an alternative, however, that comes quite close to the expression with taxare used just before our passage: namely, destinare.⁷⁴ By adding an s, one could read: secundum quod notitia subter adnexa destinet, i.e. the duces and their staffs receive the salary that the notitia (mildly personified) ‘assigns’ or ‘establishes’ for them. Justinian uses this word in a similar sense elsewhere.⁷⁵

Proposed text:

… unusquisque dux seu eorum officia, secundum quod notitia subter adnexa destinet, emolumenta sua ... percipiant.

… each commander or his staff shall, according to the amount defined in the schedule attached below, receive their perquisites ...

14. CJ 1.34.2 (143.25–6 K.)

The following passage has been restored from Basilika 6.1.103 and represents an abridgement of Justinian's original constitution. We read in Krüger's edition:

Ὁ αὐτὸς βασιλεύς. Φροντιζέτωσαν οἱ τῶν ἐπαρχιῶν ἄρχοντες καὶ αἱ τούτων τάξεις οἰκείῳ κινδύνῳ μηδεμίαν τὰ τῆς ἰδικῆς κτήσεως κτήματα βλάβην ὑφίστασθαι ἢ τὴν ἐξ αὐτῶν μειοῦσθαι πρόσοδον ἢ ἐμπαίζεσθαι. ἐξέστω δὲ τῷ κόμητι τῆς ἰδικῆς κτήσεως καὶ προστιμᾶν αὐτοῖς καὶ μετὰ τὴν ἀρχὴν τὴν γενομένην ἀνασῶσαι βλάβην, ἀλλὰ μηδεμίαν δεχέσθωσαν ἀπόρων ἢ ὁμοκήνσων ἐπιβολήν.

The same Emperor. The provincial governors and their official staffs shall, at their own peril, see to it that the property of the Imperial Patrimony suffers no loss, diminution of its revenue, or fraud. The Count of the Imperial Patrimony shall have the power to fine them and after (his?) term of office to make good the loss, and they shall not accept any addition (ἐπιβολή)⁷⁶ of unproductive or collectively assessed land.

To prevent provincial governors from diminishing imperial property, the comes sacri patrimonii ⁷⁷ is entitled both to inflict fines on them (καὶ προστιμᾶν αὐτοῖς) and to do some other thing to them for which a verb is lacking. As the text presently stands, the comes is strangely described as ‘having the power’ to make good (i.e. at his expense) any loss caused by the governors after his (or rather their?) time in office. A supplement such as e.g. καὶ προστιμᾶν αὐτοῖς καὶ <ἀναγκάζειν> ... ἀνασῶσαι κτλ. would make the governors, not the comes, subject of the second infinitive ἀνασῶσαι. Then the governor who caused the loss would have to make it up, and the administrator of the property affected (the comes patrimonii) would see to it that he did so. As with other Greek texts restored from later collections, a conservative editor may well prefer to let the text stand: other than the contents, there is no indication of how the missing infinitive could have been lost, and we are not dealing with the original text of the Codex. I incline to think that the reading in the Basilika is the result of careless editing. The subject of the next sentence, ἀλλὰ μηδεμίαν δεχέσθωσαν ἀπόρων ἢ ὁμοκήνσων ἐπιβολήν, appears to be the governors again, who were commanded in a similar third-person imperative (φροντιζέτωσαν) above.

Proposed text:

ἐξέστω δὲ τῷ κόμητι τῆς ἰδικῆς κτήσεως καὶ προστιμᾶν αὐτοῖς καὶ <ἀναγκάζειν> μετὰ τὴν ἀρχὴν τὴν γενομένην ἀνασῶσαι βλάβην, ἀλλὰ μηδεμίαν δεχέσθωσαν ἀπόρων ἢ ὁμοκήνσων ἐπιβολήν.

The Count of the Imperial Patrimony shall have the power to fine them and after (their) term of office <force them> to make good the loss, and they shall accept any addition (ἐπιβολή) of unproductive or collectively assessed land.

15. CJ 1.49.1.2 (153.28 K.)

In this final passage, the punctuation again seems in need of correction. Krüger prints the text as follows:

Nec ullam ante praefinitum tempus de provincia discedendi excusationem ei tribuat vel divina revocatoria vel codicilli alterius administrationis oblati vel praeceptum amplissimae tuae sedis, ut alterius provinciae moderatoris vices obtineat, aut praeceptum praefatae vel alterius civilis seu militaris cuiuscumque potestatis, ut quamcumque sollicitudinem publicam gerat aut exhibeatur vel deducatur, aut postremum cuiuslibet artis astutia, cuiuscumque occasionis excogitata calliditas excludatur, ut modis omnibus, quae pro universarum provinciarum salute sancimus, sortiantur effectum.

Nor may any divine letter of recall or codicils conferred on him for another office or an order from your Most Exalted Prefecture grant him any excuse for leaving the province before the appointed time, so that he might take the place of the governor of another province; nor an order of the aforesaid prefecture or another civil or military official, that he undertake some public charge or be summoned or brought elsewhere; nor, lastly, any clever scheme, any cunning devised to provide any reason whatsoever shall be eliminated, so that what We decree on behalf of the safety of all provincials shall in every way be carried into full effect.

All possible ways a governor could illicitly obtain leave from his province are forbidden. First documents originating at the imperial court (divina revocatoria and codicilli) are invalidated or banned; next a command from the praetorian prefect (praeceptum amplissimae tuae sedis) that he administer another province; third, a command from the prefect or even another official (vel alterius civilis seu militaris cuiuscumque potestatis) that the governor be given some other task, summoned or transferred out of the province; and last (postremum: that is, ‘in conclusion’), all artificial excuses (cuiuslibet artis astutia) are forbidden. A similar phrase follows – cuiuscumque occasionis excogitata calliditas excludatur – but the thought is new: the phrase qualified with postremum should really be the last item of a series. Thus cuiuslibet artis astutia numbers as the last of the things that present a governor an excuse to leave his province (Nec ... discedendi excusationem ei tribuat). In this final sentence, Justinian pauses to summarize the foregoing: every cleverly conceived excuse shall be excluded so that the decision of the emperor has the desired result.

Proposed text:

Nec ullam ... de provincia discedendi excusationem ei tribuat aut divina revocatoria ... aut postremum cuiuslibet artis astutia. cuiuscumque occasionis excogitata calliditas excludatur ...

Nor may any divine letter of recall grant him any excuse for leaving the province ... nor, lastly, any clever scheme. All cunning devised to provide any reason whatsoever shall be eliminated ...

CONCLUSION

It is useful to remember the serious obstacles that the text of the Codex Justinianus presents to potential editors and translators. Conceived as a bilingual text, the Codex Justinianus survives in separate, virtually monolingual, traditions, Latin and Greek. The text restored in modern editions is necessarily a patchwork derived from both, to say nothing about the remains of complete manuscripts.⁷⁸ Certainty about lost Greek texts, in particular, is often impossible. Both traditions are subject to the usual vagaries of textual transmission, but the extant witnesses often give no decisive, obvious answer, and in some cases it remains unclear whether a strange reading is original to the text or introduced by scribes. The language of the laws only exacerbates these problems: the imperial constitutions of the Later Roman Empire are not only legal documents that demand careful reading and interpretation, but also consciously rhetorical formulations intended to impress and persuade us. It is hoped that this article may raise awareness of the complexity of the text of the Codex Justinianus; such awareness is particularly necessary at the moment when the Codex Justinianus will be made widely available in reliable English translation. One should never grow too comfortable with a familiar edition or translation.

To conclude this small contribution to the textual criticism of the Codex Justinianus, the following emendations or supplements may be presented in a table: