African Church probably began, like the Roman, by being a Greek-speaking Church. Greek words are found in that most precious relic of the early African Church the account of the martyrdom of Perpetua; and the great African Tertullian was not only able to read Greek, but wrote some treatises in that language. Usener's discovery seemed to throw new light on the first language of the African Church; and a controversy arose: Was this Greek text of the Scillitan martyrdoms, which was confessedly older than the current Latin texts, the parent of all the Latin versions, or was it itself a translation from a lost Latin original? Usener and Hilgenfeld took the latter view; Aubé, Renan, and others took the former view, and Bishop Lightfoot, though with some hesitation, was disposed to agree with them.

This controversy was renewed by a new find. Mr. Rendel Harris made in the library of the Holy Sepulchre at Jerusalem the surprising discovery of a Greek version of the Acts of Perpetua,' which we have just mentioned, and which had been justly valued, not only as one of the most affecting and most authentic stories of a Christian martyrdom, but as one of the earliest specimens of Latin Church literature. Mr. Harris was of opinion that this Greek text was the original from which the previously known Latin texts had been translated, and he pointed out passages where the Greek cleared up obscurities or corrected what he took to be mistakes in the Latin.

We ourselves adopted Mr. Harris's opinion in a notice of his discovery which appeared in this Review at the time, for we thought that, in the absence of evidence how far the African Church was bilingual (or possibly trilingual) in the second century, we were not in a position to deny that Greek Acts might have been liturgically read in the commemoration of martyrdoms, and that we therefore were not entitled to disregard the evidence of superior antiquity which Mr. Harris found in the Greek form. Mr. Robinson, however, though admitting the Greek form to be older than the current Latin texts, was persuaded from internal evidence that it was itself a translation from an original Latin which he believed research might discover. And he verified his anticipation by himself finding older Latin texts both of the Acts of Perpetua and of the Scillitan martyrdoms.2 Mr. Harris has consequently abandoned his opinion of the priority of the Greek of the former Acts, and with regard to the latter Mr. Robinson thinks

1 He published it as a separate little volume, The Acts of the Martyrdom of Perpetua and Felicitas, London, 1890.

2 Published by him, Texts and Studies, vol. i. pt. ii.

that it is scarcely probable that the theory of a Greek original will be revived after the publication of the present Latin text.'

Now to speak first of the earlier document, the Scillitan Acts, there can be no doubt that the Greek and Latin forms are not independent. In any criticism of Martyrdoms great use is to be made of the study Les Actes des Martyrs, which Le Blant offered as a supplement to Ruinart's Acta Sincera. From the details there presented of the methods of ordinary Roman magisterial investigation we see at once that both the Greek and Latin Scillitan Acts take up the story in the middle. We do not find the ordinary opening of identification of the accused person by asking his name, father's name, country, and rank. There is no indictment and no questioning on the subject of the charge, but in both versions the report of the trial begins with the judge's offer of the emperor's pardon if the prisoners will return to a better mind. The Greek and Latin reports of the questions and answers that followed are in complete correspondence, the most important difference being that in the Greek Acts the martyrs are spoken of with the prefix ἅγιος, ὁ ἅγιος Σπερᾶτος, &c., where the Latin simply has Speratus,' &c. We cannot doubt that the Latin here more truly represents the sources whence the account was derived. The Roman courts were provided with a staff of official reporters who at trials took down the questions of the magistrates and the answers made by the accused. Christians were often able to obtain, either by favour or by purchase, these official minutes of the proceedings at the examination of their confessors, and it is needless to say that the official Acts would contain no such title as ayios. But these official minutes have not only been the basis of some of the most authentic Acts of early martyrdoms, but they have also furnished a model to which the forgers of spurious Acts have frequently conformed. And this prevents the absence of such titles from being in itself a decisive proof of superior antiquity. For instance, one of the shorter Acts of Perpetua, the comparative lateness of which is acknowledged by everybody, contains no such titles.

But the most important difference is in the conclusion. In the Latin it was, 'Et ita omnes simul martyrio coronati sunt, et regnant cum Patre et Filio et Spiritu Sancto per omnia secula seculorum. Amen.' The Greek is: TEXELONσav τῷ ξίφει, μηνὶ Ἰουλίῳ ιζ· ἦσαν οὖν ὁρμώμενοι οἱ ἅγιοι ἀπὸ Ἰσχλὴ τῆς Νουμηδίας, κατάκεινται δὲ πλησίον Καρθαγέννης μητροπόλεως· ἐμαρτύρησαν δὲ ἐπὶ Πέρσαντος καὶ Κλαυδιανοῦ τῶν ὑπάτων καὶ Σατουρνίνου ἀνθυπάτου, καθ ̓ ἡμᾶς δὲ βασιλεύ

οντος τοῦ Κυρίου ἡμῶν Ἰησοῦ Χριστοῦ, ᾧ πρέπει πᾶσα δόξα,

κ. τ. λ.

On these endings Mr. Robinson remarks, 'The close of the piece has been altered and expanded in the Greek, in which the locality of the martyrs is named, and the exquisite phrase "regnant cum Patre et Filio et Spiritu Sancto" has been rejected.' 'It is impossible to believe that the Latin could have been produced by abbreviation of the Greek.'

With regard to this we have to say that if the Greek is an expansion of the Latin it is an expansion made by some one having knowledge independent of the Latin, which contains no note of the place whence the martyrs came or of that where they were buried. Thus if the Latin be supposed to contain the form of the Martyrdom annually read in the Church of Carthage, nothing forbids us to suppose the Greek to be a contemporary document drawn up for the use of Christians elsewhere. But when Mr. Robinson speaks of this Greek as an expansion of the Latin he strangely omits to notice that the original of this Greek conclusion is to be sought not in the Latin but in the Martyrdom of Polycarp, which had been published some twenty years before, and which served as a model for many subsequent stories of martyrdoms. This Martyrdom ends with giving the date ȧv0vπατεύοντος Στατίου Κοδράτου, βασιλεύοντος δὲ εἰς τοὺς αἰῶνας Ἰησοῦ Χριστοῦ, ᾧ ἡ δόξα, κ. τ. λ. It is impossible to doubt whence the African martyrologist borrowed his contrast between the changing sway of earthly rulers and the eternal reign of Christ. Now, though we do not contest the originality of the main body of the Latin Acts, we are not without doubt as to the conclusion; for the formula 'regnante Christo,' of which Polycarp's Martyrdom affords the earliest example, came to be extensively adopted. Lightfoot 2 gives a number of examples from Ruinart's Acta Sincera, and he refers to Blondel, De Formula 'regnante Christo' in Veterum Monumentis Usu, who fills some twenty pages (pp. 371 sqq.) with instances of the use of the formula. Possibly if Eusebius had copied for us the conclusion of the Lyons Martyrdoms it might be found that they ended in the same way. The formula 'regnante' was clearly not derived from the formula 'regnant,' because we can otherwise give a good account of its origin; but it is by no means so clear that the formula 'regnant' was not derived from an original 'regnante;' and our suspicion is increased by the fact that though one of the 1 Texts and Studies, vol. i. pt. ii. pp. 116-17.

2 Ignatius, i. 636.

later Latin texts printed by Robinson bears indirect testimony to the reading 'regnant' by substituting the interpretation 'intercedunt pro nobis ad Dominum nostrum Jesum Christum,' the other has no trace at all of that reading.

The influence of Polycarp's Martyrdom is to be found also in the words immediately preceding those we have been discussing, καὶ ἀναπεμψάντων αὐτῶν τὸ ἀμὴν. Mr. Robinson finds in them a mark of lateness, 'sending up the Amen being a commonplace of martyrologists.' They are to be found in the Martyrdom of Polycarp; whether or not they are to be found in later Martyrdoms is not very relevant. But if the introduction of these words is a commonplace with martyrologists, may it not be because the fact which they record was common? We know from many sources that there was no more striking feature of early Christian worship than the loud-sounding Amen with which the congregation adopted as their own the words of the leader of their devotions. Who can find anything incredible in the statement that the martyrs employed in prayer the last few moments while their sentence was being proclaimed by the herald, and before its actual infliction? Doubtless many who could not hear the words of their prayer could hear the loud Amen with which the sufferers assented to the prayer of their chief; and if the same thing is told of other martyrdoms, may it not be because the same thing often occurred? However we may put aside the questions whether there is evidence that the Scillitan martyrs uttered no words beyond the 'Deo gratias' which is all that the Latin text records; whether it was impossible for any bystander to have heard the actual words uttered by Speratus, and to have furnished them to the Greek martyrologist. Let it be granted that the prayer of the Greek text was pure invention after the model of Polycarp's prayer, and the question remains for the critic, Is there anything improbable in the supposition that this version of the martyrs' prayer may have been given while the event was still recent? And on the whole it seems to us a probable conclusion that the Greek text represents the account of the Scillitan martyrdoms which the Carthaginian Church sent at the time to distant Churches.

We come now to the Acts of Perpetua, but premise that our judgment about them must not be biassed by any we may have formed about the Scillitan Acts. The Acts now to be considered are twenty years later, and much may have occurred in the meantime. They embody two documents of surpassing interest, viz. a history by Perpetua herself of what had befallen

her from the time of her first apprehension until the day before her martyrdom, and an account by a fellow-martyr, Saturus, of visions he had seen while in prison.

We have to protest against an abuse of language into which Mr. Robinson has been led by Mr. Harris, an abuse likely to lead both to critical and chronological mistakesnamely, the use of the word Montanist without any evidence that the persons to whom it is applied had ever heard of Montanus, merely on the ground that they believed God's Holy Spirit was ever working in His Church, or because they did not hold that the age of miracles had passed, and did not regard it as impossible that wonders such as had taken place in old time might manifest themselves in their own day. If this is Montanism the whole Early Church was Montanist, and it is not too much to say that the numerical majority of Christians are Montanist at the present day. Of course there is a great difference between admitting the possibility of supernatural manifestations and believing that in a particular case they occurred. Many a sober-minded Roman Catholic who does not believe that the Blessed Virgin appeared at Lourdes, and knows that his Church does not require him to believe it, would yet be sorry to say that a story of such an appearance might be rejected as in itself incredible. Christians in the third century were as free as Roman Catholics are now to form their own judgment with respect to alleged miraculous manifestations. And a difference of opinion on such a subject, there being no doctrinal difference, would have been no cause of separation if a question of ritual had not intervened. It was the insisting, as obligatory on all Christians, on fasts and other ordinances resting solely on the authority of Montanist revelations which made the interference of Church authority necessary; and the Roman Church adopted the judgment of the local Phrygian Church, which was unfavourable to the Montanist pretensions. After that Montanism became schismatic in the West, but it had not been so before.

Now it is certain that Perpetua and Saturus believed that they had themselves been favoured with visions and revelations; and it may be presumed that if they were told of similar manifestations as exhibiting themselves in a distant part of the Church they would be well disposed to believe the story, and so no doubt would be many other pious Christians in Africa. And there would be nothing schismatical in their holding this belief as long as no contrary decision had been pronounced by the Church. We may count it as certain that no such decision had been pronounced at the time of Per

VOL. XXXVI.- NO. LXXI.

E