NON-CANONICAL BOOKS

Having in Lectures xi. and xix. spoken of Apocryphal Gospels and Apocryphal Acts, I now add a lecture on other books known to the early Church, but which did not find admission into the Canon.

The Apocalypse of Peter. I give the first place to this work, because it claimed Apostolic authority, and because we infer from the Muratorian Fragment (see pp. 50, 225), that it had obtained a place, though not an undisputed place, in Church reading before the end of the second century. With regard to its contents we have only positive information as to two passages, both indicating that the book contained a description of the Last Judgment. One of these is preserved by Clement of Alexandria in the Prophetic Selections (41, 48), which, according to the general opinion of scholars, formed part of his Hypotyposeis. Clement, who is habitually indiscriminate in his reception of books, cites this Apocalypse as a genuine Petrine work¹ and as Scripture; but the extract which he preserves gives us no favourable opinion of it. It deals with the future condition of abortive births, and of children born in adultery, exposed by their parents. The former, it says, will be handed over to an angel nurse (ἀγγέλῳ τημελούχῳ), under whom they will receive instruction, and after suffering what they would have suffered if they had lived in the body, will attain the better abode. The exposed children receive like nursing and instruction, and grow to the condition of the faithful here of the age of a hundred. On account of the injustice done them they obtain mercy and salvation, but only so far as freedom from punishment. I should infer that the writer must have held the general necessity of baptism in order to salvation, a special exception being made in favour of these murdered infants, who, it may be remarked, were presumably the children of heathens. The passage goes on to tell that the bright shining of these children shall strike like lightning the eyes of their unnatural mothers, from whose unused milk shall be generated carnivorous little beasts which shall devour them. I have quoted these puerilities at length, because the passage furnishes proof that the Apocalypse of Peter retained high consideration so late as the beginning of the fourth century. Methodius (see p. 393) says: We have received in the divinely-inspired Scriptures, that even those who are begotten in adultery are handed over to angel nurses (τημελούχοις ἀγγέλοις). For if they came into being in opposition to the will and decree of the blessed nature of God,² how should they be delivered over to angels to be nourished with much gentleness and indulgence? and how could they boldly cite their own parents, before the judgment-seat of Christ, to accuse them, saying: " Thou didst not, O Lord, grudge us thy common light, but these exposed us for death, despising thy command "? (Sympos. ii. 6). There can be no doubt that what Methodius here cites as divinely-inspired Scripture is taken from the passage of Peter's Apocalypse that is quoted by Clement of Alexandria.

The other extant passage of this Apocalypse is pre served by Macarius Magnes (see p. 163). We can infer that at the very end of the fourth century it had not quite lost its consideration. The heathen objector, as if the book were recognized by Christians as an authority, selects a saying of it for attack The earth shall present all to God in the Day of Judgment, and itself shall then be judged with the heaven that surrounds it. Macarius,³ in reply, remarks that it will not avail him to decline the authority of that Apocalypse, the same doctrine being taught in Is. xxxiv. 4, and Matt. xxiv. 35.

I quoted (p. 476) the formal judgment of Eusebius (iii. 25) about this book. He places it with the Epistle of Barnabas, and the Shepherd of Hermas in the second rank of disputed books (which he calls νόθα), or books not canonical, but known to most ecclesiastical writers, and which stand on a different level from books of heretical origin (among which he names the Gospel of Peter), which no ecclesiastical writer has deemed it fit to make use of. In an earlier passage (iii. 3) Eusebius has with less discrimination lumped together all the Apocryphal books ascribed to Peter (the Gospel of Peter, the Acts of Peter, the Preaching of Peter, and the Revelation of Peter), as not received among Catholics, no ecclesiastical writer either of former days or his own having used testimonies from them. We have seen that the last sentence is too strongly worded, as far as the Apocalypse of Peter is concerned; but there can be no doubt that Eusebius is, in the main, right as to the weakness of external attestation for the book. And that it had generally dropped out of Church reading in his time may be inferred from his classing it not with the minor Catholic Epistles, but with the Epistle of Barnabas and the Shepherd of Hermas. But a hundred years after the death of Eusebius its use was not absolutely extinct; for Sozomen, in speaking (vii. 19) of singular local usages in different Churches, tells that in his time this Apocalypse, though regarded as spurious by the ancients, was still annually read on Good Friday in some Churches of Palestine. Its continuance for some time in Church use is also testified by its being included in the Stichometry of Nicephorus (see p. 178), where it immediately follows the Revelation of St. John, and in the list of the Codex Claromontanus (see p. 473). Both these authorities agree in making the length of the book something less than a quarter of that of the Apocalypse of St. John, the number of στίχοι being in the former list 1400 and 300, respectively; in the latter 1200 and 270. It has even been conjectured that this had originally formed part of the Sinaitic MS., of which six leaves have been lost, coming between the Epistle of Barnabas and the Shepherd of Hermas. These leaves, no doubt, contained one of the disputed books; and the Revelation of Peter is not too long to have been included in them. But it is doubtful whether it was long enough to fill the gap, and Mr. Rendel Harris (Johns Hopkins University Circulars, 1884, p. 54) has urged the preferable claims of the Psalms of Solomon,⁴ which originally followed the Canonical books in the Alexandrian MS. Each page of the Sinaitic ordinarily contains four columns; but the poetical books of the Old Testament are written in στίχοι, or verses divided according to the sense, and with only two columns on a page. Now, the Epistle of Barnabas ends on the third column of a page, and the fourth is left blank, contrary to the scribe's usual practice. This would be explained, if the book which was immediately to follow was poetical, requiring two columns on a page. Thus, the book of Malachi ends on the third column of a page, and the fourth is left blank, because the following book (the Psalms) is written στίχηδόν.

It is barely worth while to mention conjectural attempts to discover traces of the influence of Peter's Apocalypse. The extant fragments of the treatise on the universe, by Hippolytus, contain a description of the unseen world and the intermediate state, which Bunsen imagined to have been derived from this source. With less probability, Hilgenfeld claims for this Apocalypse a passage twice quoted by Hippolytus (De Antichrist., 15, 54) as a saying of a prophet, but not found in our text of the Old Testament. It is not likely that Peter would have been cited as the prophet; and, not to quote other instances, we have seen (p. 478) that early Fathers sometimes read in their Old Testament text passages not found in ours. From the assumption, however, that the prophet means the Apocalypse of Peter, Hilgenfeld draws a startling inference. He finds further on (c. 68) in the same treatise of Hippolytus: The prophet says " Awake thou that sleepest, and arise from the dead, and Christ shall give thee light; " and he concludes that the original of this saying is also to be traced to Peter's Apocalypse, whence it was borrowed by the author of the spurious Epistle to the Ephesians! Hilgenfeld's discussion is to be found in the last fasciculus of his Nov. Test. ext. Can. recept., 2nd edition, 1884.

I will not speak at length of other Apocalypses, none of which can be called really early. The most important is that of Paul, whose account, 2 Cor. xii. 2-4, of the revelations with which he had been favoured offered a temptation to a forger to atone for the Apostle's silence on the subject. Accordingly we hear from Epiphanius (xxxviii. 2) that the Gnostics had an ἀναβατικὸν Παύλου, which professed to be a secret record of the mysteries then revealed to the Apostle. All trace of this book has been lost. That which has actually come down to us as the Apocalypse of Paul is much later. Sozomen, in a passage (vii. 19) already cited, tells that a work thus inscribed was in much esteem among the monks, and he reports that the book was said to have been found by divine revelation in the reign of the then present emperor (Theodosius the younger) buried in a marble box, under what had been the house of Paul at Tarsus. Sozomen ascertained from an aged presbyter at Tarsus that this story was not true. The same Apocalypse is condemned by Augustine (in Johan. Evang. c. 16, tract. 98). It is to be found in Tischendorfs Apocalypses Apocrypha (1863), an d more recently has been the subject of an investigation by Brandes, Visio Pauli, 1885. I content myself with mentioning that the appearance in the book of an angel Temeluchus indicates that the author had studied the Apocalypse of Peter.

The Epistle of Barnabas. A second work included by Eusebius in his list of disputed books bears the name of a member of the Apostolic company, the Epistle of Barnabas. It is found in the Sinaitic MS., beginning on the leaf where the Revelation ends, and placed, together with the Shepherd of Hernias, as a kind of appendix to the New Testament books. Its being found at all in a MS. intended for Church use seems to indicate that it had at one time been used in the public reading of the Church, while its position at the end shows that at the time the MS. was written it stood on a lower level than the Canonical writings. The same thing may be inferred from its inclusion among the antilegomena in the Stichometry of Nicephorus, where it follows the Revelation of Peter. It is quoted several times by Clement of Alexandria,⁵ who calls its author sometimes the Apostle Barnabas, sometimes the Prophet Barnabas. Else where he states that he was one of the Seventy; and one passage is worth quoting as throwing light on the authority which Clement ascribed to the Epistle. It is taken by Eusebius (ii. 1) from the seventh book of the Hypotyposeis: Our Lord after his Resurrection communicated the Gnosis to James the Just, John, and Peter: these communicated it to the other Apostles, and the other Apostles to the Seventy, of whom Barnabas also was one. Accordingly, Clement would regard the Gnosis, of which the Epistle under consideration is full, as really a divine tradition, though only reported second-hand. Origen also appeals to the Catholic Epistle of Barnabas (Adv. Cels. i. 63), and cites it as Scripture (Comm. in Rom. i. 24). These two Alexandrian witnesses make up nearly the whole of the testimony favourable to the Epistle. If it were not for the existence of an early Latin translation, we might even doubt whether it was known at all in the West before the fourth century. One coincidence with Justin and Irenaeus has been mentioned (p. 535); but in another place that admits of comparison, an allegorical interpretation of the law concerning clean and unclean animals, Irenaeus (v. 8) seems to be quite independent of Barnabas ( i o). Tertullian (Adv. Marc. iii. 7) appears to be clearly indebted to Barnabas (7) in describing the scapegoat as pierced and spit upon; yet, if he knew our Epistle as that of Barnabas, it seems strange that he should ascribe the same authorship to the Epistle to the Hebrews. Jerome (De Viris Illust. 6: see also Comm. in Ezek. xliii. 16) makes no doubt that the author of the Epistle was the Barnabas of the New Testament, but says that the Epistle is counted among apocryphal Scriptures. Elsewhere (Dial. cont. Pelag. iii. 2) he quotes from the Epistle a saying which had been previously quoted by Origen (Adv. Cels. i. 63); but he attributes it to Ignatius, probably through lapse of memory.

Turning to the internal evidence, we find the contents of the book such as certainly would not make us wish to include it in our Canon of Scripture. To cite one oft-quoted pas sage, Barnabas misquotes the book of Genesis (see Gen. xiv. 14; xvii. 27), as recording that Abraham circumcised 318 of his household, a number expressed in Greek by the letters τιή. It does not appear whether Barnabas called to mind that the book had been written not in Greek but in Hebrew. At all events he expounds that ιή denote Jesus, and τʹ the cross; and he is so satisfied with his exposition that he adds, No one has received a more genuine word from me than this; but I know that ye are worthy. ⁶ He goes on to ex plain the meaning of the prohibitions against eating the flesh of the animals counted as unclean, of all of which he gives spiritual explanations, in which the natural history is quite as curious as the theology. These spiritual explanations constitute the Gnosis which, in the mind of this author, gives him his chief claim on his readers attention. One example will suffice. The prohibition to eat the hyena means that we are to avoid adultery and other such sins; for this beast changes its sex each year, being one year male, the next female. I remember that when I was a young student myself I heard some of these passages quoted in a sermon in our chapel by one whose memory we still hold in honour. The preacher's view was that the Epistle was a genuine work of the Apostle Barnabas, and he produced the passages in order to show what rubbish an Apostle was capable of writing when he was not inspired. He thought thereby to exalt the authority of the inspired Scriptures as being sui generis, and unlike not only the writings of other men, but the writings of the same men when not inspired. His object was to establish the supreme authority of Scripture, but in real truth he did just the reverse. For according to this view the authority of Scripture must yield to whatever authority it is that settles which of the Apostolic writings are inspired, and which not. I own I know no proof that the Apostles were inspired in a different way when they were writing and when they were speaking; and in a different way when they were writing some books and when they were writing others. And, as I have said, if this view be correct, the supreme authority in the Church is that which brings Apostles to its bar, tests their writings, and assigns to some the attribute of inspiration which it denies to others. But what that authority is I don t know. I know that the general sense of the Christian Church has refused to put the Epistle of Barnabas on a level with those of St. Paul; but if you ask by what tribunal, or by what formal act this conclusion has been arrived at, I should be as much puzzled as if you asked me by what tribunal it has been decided that Shakespeare is a greater poet than Beaumont and Fletcher. Without saying anything about the Church's claim to expect Divine guidance, we can hardly refuse to yield at least as much deference to her decisions as we pay to received opinion in matters of taste. And so, no matter who wrote the Epistle we are considering, we shall not accept it as inspired. But if we believe the Apostle Barnabas to have been the author, since he was a man who in his lifetime had claims, like those of St. Paul, to be God's inspired messenger, we require a theory to explain the grounds on which we are to maintain that the writings of one are more above our criticism than those of the other.⁷

It is perhaps not preparing you to judge with quite unbiassed minds of the question of the authorship of the Epistle that I have allowed you to see what consequences are likely to follow if the Apostolic authority be conceded. But judges who are above being prejudiced by considerations of this sort, and who would have no difficulty in believing Apostles to have been guilty of any amount of error, have pretty unanimously decided that the Epistle was written at a later time than Barnabas is likely to have lived to, and that the author is a different manner of man from what the historical Barnabas is described as having been. The main argument is derived from the whole attitude of the writer towards Judaism. The historical Barnabas was a Levite, and was trusted by the Jerusalem Church, to whom he introduced Paul. In his only difference with St. Paul on the subject of Judaism he erred by too great concessions to the Jewish party. Now the writer of the Epistle does not show that acquaintance with Jewish rites which the Levite Barnabas must have had. I exemplified to you, in the case of the number 318, that he does not quote the Old Testament accurately. In fact gross inaccuracy is the rule with him; and in his account of Jewish rites (and on the symbolizing of Christ by these rites he builds many arguments) he deviates widely from the Old Testament. Nor can we have recourse to the supposition that the rites traditionally practised in Jerusalem at that time differed from those prescribed in the Old Testament; for the Talmud, which may be supposed to have preserved Jewish traditions, gives the so-called Barnabas as little countenance as the Old Testament does.

But more remarkable even than his inaccuracy in speaking of Jewish institutions is his total want of respect for them. He does not look on the performance of the Jewish rites as introductory and preparatory for Christ, but as a gross sin a misconception of the true meaning of the law. He has a spiritual exposition for the Mosaic precepts, and he holds that the Jews, by taking them literally, excluded themselves from God's covenant. He even represents the Jews as deceived by an evil angel. Paul forbade the Gentiles to be circumcised; but, in Acts xxi., the statement is repelled as a calumny that he taught the Jews to forsake Moses, and not to circumcise their children nor walk after the customs. This writer, under the name of Barnabas, would seem to condemn the Jews for having observed such customs even before our Lord's coming. And his whole tone of feeling towards the Jewish nation is such, that when I balance the probabilities that a born Gentile should acquire as much knowledge of the Old Testament as this writer displays, or that a born Jew should come to feel towards his own nation so completely as an outsider, I prefer to embrace the former probability.⁸

A less formidable difficulty in the way of ascribing the authorship to the Apostle Barnabas arises from the date of the Epistle. There is a range of some forty or fifty years within which the date may lie; but it is certain (ch. 16) that it is later than the destruction of Jerusalem. Now (see p. 466) we should not expect to find the Apostle Barnabas in activity so late; and the silence of Paul's later Epistles about him might lead us to think he had died before Paul. But this is only a presumption which must yield to any good evidence on the other side; and Paul's silence would be accounted for if Barnabas had gone off to work in a completely different sphere for ex ample, Egypt. A limit in the other direction to the date of the Epistle is furnished by its complete silence as to any of the Gnostic theories which caused so much controversy in the Church quite early in the second century. The anti-Judaism of the Epistle might make us think of Marcion; but the Epistle is distinctly pre-Marcionite, there being not the least trace of any of the notions peculiar to that heretic.⁹ On these grounds the Epistle cannot be dated later than A. D. 120. There are two passages which have been used to determine more precisely the date of the Epistle. In ch. 4, in proof that the last days are at hand, he quotes Daniel's prophecies (vii. 8, 24) often kings, and of one king overthrowing three others. He does not enter into the question how the ten kings were to be made out, but merely remarks, ye ought therefore to understand. The brevity of this comment indicates that Barnabas found the fulfilment of the prophecy in some patent fact, and not in one requiring historical or chronological studies to discover it. I therefore know no explanation of his words so natural as that the Epistle was written in the reign of Vespasian. It is true that a historical student might discover that, counting Julius Caesar, Vespasian was only the tenth emperor, while Daniel's words would lead us to think of his little horn as representing an eleventh king; but Barnabas is one of the last writers from whom minute accuracy of interpretation need be expected. If he lived in the reign of Vespasian, the rapid overthrow in succession of three emperors, Galba, Otho, and Vitellius, might naturally make him think that he was witnessing a fulfilment of Daniel's prophecy of one king subduing three. I know no other time when his language would be natural. On this account, though some other considerations would induce me to push down the date of the Epistle to the second century, I find it hard to resist the inference that we must ascribe it to the reign of Vespasian, A. D. 70-79. In the other passage (16) he quotes the prophecy They that destroyed this temple shall themselves build it up again,¹⁰ and adds, and so it comes to pass. Through their making war it was destroyed by their enemies; and now both they and the servants of their enemies shall build it up again. It has been supposed that this refers to some attempts to rebuild the Temple in the reign of Hadrian; but I find no evidence of anything of the kind to give a probable explanation of the language of Barnabas; and it seems to me plain from the rest of the chapter that it is in the building up of a spiritual temple that he finds the fulfilment of the prophecy. The argument, therefore, for the earlier date, drawn from the former passage, remains undisturbed.

There is nothing in the letter itself to determine the place to which it was addressed; but since it is from Alexandria we first hear of it, it seems probable enough that it was sent to that city. Alexandria contained a large Jewish population, and thus the conflict with Judaism would there occupy much of Christian attention. Possibly, too, some Jewish rites may have been different in Egypt and in Palestine. The name Barnabas, found in the title of the letter, does not appear in the letter itself. All that we discover from it is that it was written by a Christian teacher to a Church in which he had himself laboured, and to which he was accordingly well known. We are not forced to suppose that it was written from a distance: the author may have merely wished to leave his people a written record of his teaching. If the author was not the Apostle Barnabas and I find it hard to believe he was the question will be asked how the letter came to bear his name. The best conjecture I can make, setting aside the guess that the author's name may really have been Barnabas, is that the Church of Alexandria was founded, if not by Barnabas himself, by men of Cyprus, who owed their knowledge of the Gospel to him, and that so his name came to be attached to a venerable record of early teaching preserved in that Church.

The Epistle of Clement. This venerable document has clearly a right to be next considered. It is true that although Eusebius Calls the Epistle μεγάλη, θαυμασία, ἀνωμολογημένη παρὰ πᾶσιν (iii. 1 6, 37), he does not include it in his list of ecclesiastical books (see p. 475); and even if the omission arose from inadvertence, the possibility that the book could be forgotten shows that it had no serious pretensions to canonical authority when Eusebius wrote. But it had evidently made a profound impression on the earlier Church. It was written in the name of the Church of Rome¹¹ to the Church of Corinth, and was intended to appease a sedition in the latter Church, ending in the unwarrantable deposition of some presbyters from their office. The letter, which is framed on the model of the Apostolic Epistles, is mainly taken up with enforcing the duties of meekness, humility, and submission to lawful authority. The reception it met with in the Church to which it was addressed is evidenced by a letter written about A.D. 170, by Dionysius, then bishop of Corinth, to Soter, bishop of Rome, to acknowledge a gift of money which the Roman Church had sent, exercising their hereditary custom of liberality. Dionysius states that the letter accompanying this gift had been read at their meeting on the Lord's Day, and would continue to be so read for their edification, as also the former letter of the Roman Church, written by Clement (Euseb. iv. 23). The public reading of Clement's letter spread to other Churches; and Eusebius (iii. 16) says that he knew of the practice existing in very many Churches, both formerly and in his own time (see also Jerome, De Viris Ill. 15, Photius, Cod. 113). With this agrees the fact that it is found (together with a second Epistle) in the Alexandrian MS. of the New Testament, but coming as a kind of appendix after the Apocalypse. The scribe, however, has included it among New Testament books in his table of contents; and in a Syriac version, to be mentioned presently, it is even joined to the Catholic Epistles. On the other hand, in the list of Nicephorus it is not even placed with the antilegomena in company with the Apocalypse of Peter and the Epistle of Barnabas, but among the Apocrypha, with the Acts of Peter, John, and Thomas. It seems to have been scarcely known to the Western Church, and there is no evidence of any early translation into Latin. The second-century attestation to the Epistle is copious. It is clearly referred to by Hermas in a passage which will come under consideration in the next section; it is recognized by Hegesippus (Euseb. iii. 16, iv. 22), who speaks of it in connexion with his visit to Corinth, and probably found it in use there; it is cited by Irenaeus (iii. 3), and several times by Clement of Alexandria, who once (Strom, iv. 17, p. 609) gives Clement the title of Apostle, and another time (vi. 8, p. 272) cites by mistake a passage of Clement as from the prophet Barnabas. Probably Clement found the two Epistles of Clement and Barnabas together,, appended to his Apostolus or collection of Apostolic letters. But the impression made by Clement's revival of the Apostolic method of teaching distant Churches is testified even more strongly by the indirect evidence of the use made of his letter. It is a matter of dispute whether certain coincidences in the Epistles of Ignatius are sufficient to prove acquaintance with Clement's letter, but there can be no doubt as to the constant employment of it in the Epistle of Polycarp. The beginning and ending of the letter of the Church of Smyrna, relating the martyrdom of Polycarp, are both fashioned after the pattern of Clement's Epistle; and his form of address, the Church sojourning in Rome (παροικοῦσα Ῥώμην) to the Church sojourning in Corinth, became an established formula, which was adopted in the letters of Dionysius of Corinth (Euseb. iv. 23), and of the Churches of Vienne and Lyons (v. i). And further evidence is furnished by the legendary stories, having Clement for a leading personage, which gained so much circulation by the end of the second century or the beginning of the third. There can be no doubt that it was the celebrity which his widely-circulated Epistle had given to the name of Clement which recommended that name to the inventors of these legends.

The letter begins by explaining that it would have been written earlier if it had not been for repeated calamities in which the Church of Rome had been involved. It used to be supposed that the persecution under Nero was here referred to, but the best critics are now agreed that all the notes of time in the letter oblige us rather to refer it to the reign of Domitian,¹² during which the Roman Church had to suffer a severe trial of persecution. The date would thus be about A.D. 96. This date well agrees with the statement of Irenaeus (iii. 3), probably derived by him from Hegesippus, that the Apostles Peter and Paul, having founded the Church of Rome, committed the government of it to the Linus who is mentioned in the Epistle to Timothy; that to Linus succeeded Anencletus, and to Anencletus Clement. Thus Clement is separated by two Episcopates from the time of the Apostles. This corresponds very well with the interval between the reigns of Nero and Domitian, but cannot be reconciled with the fiction which made Peter first bishop of Rome, and Clement his immediate successor. When this fiction came to be accepted as historical truth, it was attempted to mend the chronology by a theory that Linus only held office as Peter's deputy, and dying during that Apostle's lifetime, was succeeded by Clement; Anencletus, who has left no mark on history, being degraded to the third place. But there is every reason for adhering to the account of our oldest witness, Irenaeus. The names Linus, Cletus, Clement, have from the earliest times been commemorated in that order in the Roman Liturgy. What inducement could there have been for thrusting the unknown name of Cletus before that of Clement, unless it had a chronological title to precedence? If we have found reason to think that Clement belongs to the reign of Domitian, we cannot attach much value to a guess of Origen's (In Johann. i. 29), that he was the same as the Clement mentioned by Paul (Phil. iv. 3). The name is far too common a one to allow of our disregarding the difficulties of place and time which stand in the way of an identification.

In modern times it has been imagined that Episcopacy had not arisen before the end of the first century, and that Linus, Cletus, Clement, were but the names of leading presbyters. But if so, we may ask, how came it that the letter of the Roman Church should be universally known as the letter of Clement, whose name is not once mentioned in it? I know no good explanation of this but the old one, that this was because Clement was generally known to be at the head of the Roman Church at the time the letter was written. We need not suppose, however, that the name bishop was then distinctively used to denote the head of the Church, nor are we bound to think that the line of separation between him and other presbyters was as marked as it became in later times. The words bishop and presbyter are used inter changeably by Clement, as in Paul's Pastoral Epistles. It has been thought, however, that although Clement's letter exhibits the prominence of a single person as chief in the Church of Rome, it affords evidence that there was no such prominence in the Church of Corinth, whose bishop is not mentioned in the Epistle. But this inference is not warranted; for it is plain from the letter itself that if Corinth had ever had a bishop, he was out of office at the time the letter was written. The letter was occasioned by the deposition of certain presbyters; and it has been just said that Clement would use the name presbyter in speaking of what we now call the bishop. Now, it is to be observed that the state of things at Corinth is not adequately described by such phrases as schism, feuds, dissensions. Clement calls it (ch. i) an abominable and impious sedition (μιαρᾶς καὶ ἀνοσίου στάσεως), which he compares (ch. 4) to the sedition which Dathan and Abiram made against Moses.¹³ Accordingly he does not attempt to heal the Corinthian schism by exhortations to mutual con cessions; but he rebukes those whom he addresses, and exhorts them to unequivocal submission to the authority which they had resisted. He tells them of the necessity of order in things temporal and in things spiritual; he tells them that those whom they had deposed held an office instituted by, and handed down from, the Apostles themselves. And he says: It is shameful, dearly beloved; yes, utterly shameful and unworthy of your conduct in Christ that it should be reported that the very steadfast and ancient Church of Corinth, for the sake of one or two persons, maketh sedition against its presbyters. Ye, therefore, that laid the foundation of the sedition submit yourselves unto the presbyters, and receive chastisement unto repentance, bending the knees¹⁴ of your heart. The letter throws no light on the question whether the presbyters deposed were all equal in rank, or whether one was superior to the rest.

It bears on the question of Roman supremacy that we should understand the amount of disorder in Corinth. If there had been merely a schism there, we might wonder that Rome should undertake to arbitrate between rival claimants to office in a distant city. But if it be under stood that the Corinthian Church had distinctly violated what was elsewhere recognized as Apostolic order, the letter ceases to give evidence of Roman supremacy, for the enormity of the offence would give to a distant Church the right of expostulation. Clement's language: If certain persons should be disobedient unto the words spoken by Him through us, let them understand that they will entangle themselves in no slight transgression and danger; but we shall be guiltless of this sin, does not appear to me to indicate any official superiority of his Church, but only to be such as any Christian preacher might use in rebuking known sin. No Church was better entitled to use expostulation with another than the Church of Rome, which exercised liberality towards the rest, not only in hospitable treatment of the strangers whom business was continually drawing to the great capital, but also, as we have just seen, in direct gifts to foreign Churches. But, no doubt, this early example of successful interference must have done much to increase the prestige of the Church by whose exertions peace had been restored.

In Clement's Epistle such copious use is made of the Old Testament, that it may be probably inferred that the author was a Jew by birth, familiar with the book from childhood. In citing it the ordinary formulae of Scripture quotation are used; but the books of the New Testament are treated differently. Clement shows his acquaintance with them by weaving their language into his discourse; but he does not formally quote them as authoritative Scripture, except that he uses in this way sayings of our Lord, which, however, would seem in his use of them to derive their authority from having been spoken by Him, rather than from the book in which they were recorded.

Until lately Clement's Epistle had been only preserved in one MS. (viz., as already stated, the Alexandrian MS. of the New Testament): and there not complete, for a leaf of this part of the MS. had been lost. But a few years ago notice was taken that a manuscript book in a library at Constantinople contained, among other early writings, a copy of Clement's Epistles. Its text was made known to scholars, in 1875, in an admirable edition of Clement, published by Bryennius, metropolitan then of Serrae, now of Nicomedia, a prelate whose learning does honour to the Church to which he belongs. And, strange to say, almost about the same time a third authority for the text was recovered in a Syriac version, contained in a Syriac N. T. acquired by the University of Cambridge. In this MS. Clement's Epistles regularly take the place of New Testament books, coming, as part of the Catholic Epistles, after Jude, and before the Pauline Epistles, and even furnish ing lessons for Church reading.

Although I professed to treat of the Epistle of Clement, I have just used the plural number, Epistles, for our MS. authorities give us two Epistles ascribed to Clement. Eusebius, who usually speaks of Clement's Epistle in the singular number, mentions (iii. 38) that there was a second Epistle which bore Clement's name, but that it had not as much circulation as the former, and that it had not been quoted by the ancients. And internal evidence shows that the second, though an early document, is later, by at least a generation, than Clement's genuine Epistle. Indeed, now that we have the document complete (for the mutilation of the Alexandrian MS. had until lately deprived us of the conclusion), we learn that it is not an Epistle at all, but a written homily intended to be publicly read in Church. The writer is distinctly a Gentile, and contrasts himself and his readers with the Jewish nation in a manner unlike the genuine Clement. And instead of confining his quotations to the Old Testament, he has many citations from the Gospels, giving in one place the name Scripture to the source of his quotation. He used Apocryphal Gospels besides: one of his quotations we can trace to the Gospel according to the Egyptians. Yet he appears to have written before the great conflict with Gnosticism began, so that we may confidently ascribe the document to the first half of the second century.

The Shepherd of Hermas. Returning now to Eusebius's list of disputed books, I come to treat of the Shepherd of Hermas. The passage quoted from the Muratorian Fragment (p. 50) testifies the high consideration in which the book was then held. Although the writer refuses to the Shepherd a place in public Church reading, he lays down that it not only might, but ought to be read in private, and his language plainly indicates that, in some places at least, the Church use of the book had been such as to cause danger of its being set on a level with the Canonical Scriptures. Irenaeus (iv. 20) actually quotes a passage from the book, with the words Well said the Scripture. Clement of Alexandria quotes the book several times, and to all appearance fully accepts the reality and divine character of the revelations which it contains. Origen, commenting on Rom. xvi. 14, says: I think that this Hermas is the author of the book which is called the " Shepherd," a writing which seems to me very useful, and, as I think, divinely inspired. But his references to the book elsewhere clearly indicate that it did not then stand on the level of the Canonical Scriptures; and he several times owns that it was not received by all.¹⁵ In fact, the rise of Montanism made the Church much more cautious in the use of non-Canonical writings. It was felt that the prerogatives of Scripture were infringed on, when the utterances of modern prophets were circulated as having like claims on the acceptance of Christians. An opponent of the Montanists (Euseb. v. 16) declares that he had abstained from writing against them, lest he should seem to desire to add anything to the word of the Gospel of the New Testament, to which no one who is resolved to walk according to the Gospel can add anything, and from which he cannot take away. This state of feeling led to a severer scrutiny of the claims of books which had been admitted into public Church use; and it is intelligible why the Muratorian writer should deprecate the Church use of a book which he believed to be not more ancient than the Episcopate of Pius. The change of feeling as to Hermas took place in the lifetime of Tertullian. In an early treatise (De Oratione] he disputes against certain persons who thought themselves bound to sit down at once after prayer, because Hennas is recorded to have done so. The book must evidently have enjoyed high authority when its narrative statements could thus be turned into rules of discipline. Tertullian, in reply, says nothing to disparage the authority of the book, but only contends that such an inference from it is not warranted. That the book then existed in a Latin translation maybe inferred from Tertullian's describing it by its Latin name, Pastor contrary to his practice in speaking of books which he knew only in Greek. In a work written several years later, and after the rise of Montanism (De Pudic. 10), Tertullian contemptuously repudiates the authority of the Shepherd, declaring that it was not counted worthy of being included in the Canon, but had been placed by every Council of Churches, even of the Catholic party, among false and apocryphal writings.¹⁶ But that the book still continued to enjoy some consideration appears from Tertullian's going on to speak (c. 20) of the Epistle to the Hebrews as more received in the Churches than that apocryphal " Shepherd" of the adulterers. It is worth while to copy what Eusebius says of the book (iii. 3): It is to be observed that this book has been disputed by some, on whose account it cannot be placed among the homologoumena; but by others it has been judged most necessary for those who have especial need of elementary instruction. Hence, also, we know that it has been publicly read in Churches, and I observe that some of the most ancient writers have employed it.¹⁷ With regard to what is here said about introductory instruction, it is to be remarked that the feeling grew up that the books of Scripture were the property of the Church, and therefore could not so fitly be used in teaching those who had not yet been admitted to it. And so Athanasius (Ep. Fest. 39) classes the Shepherd, with the teaching of the Twelve Apostles and with some of the deutero-canonical books of the Old Testament, as not canonical, but useful to be employed in catechetical instruction.¹⁸ The Shepherd forms part of the appendix to the Sinaitic MS.; it is also included in the list of the Codex Claromontanus, and some twenty Latin MSS. survive to attest that it had some circulation in the West.

The book, the history of whose reception I have sketched, consists of three parts. The first part, called Visions, relates different revelations with which the author had been favoured, stating particularly the occasion and place of receiving each vision. The scene of each of these visions is laid in Rome or its neighbourhood, so that the document clearly belongs to the Roman Church. This part concludes with a narration of the vision which gives the name to the book. A man comes to Hermas in the garb of a shepherd, and tells him that he is the angel of repentance, and that he has come to dwell with him, being the guardian to whose care he has been intrusted. This Shepherd then gives him, for his own instruction and that of the Church, the Commandments, which form the second, and the Similitudes, which form the third part of the work. With regard to the general purport of these revelations, it will suffice here to state briefly that they are intended to rebuke the worldliness with which the Church had become corrupted; to predict a time of great tribulation as at hand, in which the dross should be cleared away, and to announce that there was a short intervening time during which repentance was possible, and would be accepted. The question as to the possibility of forgiveness of post-baptismal gross sin was then agitating the Church. The solution which Hermas offers is, that during that short respite the then members of the Church might obtain forgiveness. But only once: for this was an exceptional favour, and those who joined the Church afterwards must expect no other forgiveness than that which they obtained in baptism.

Concerning the date of the Shepherd, received opinion still accepts the statement of the Muratorian Fragment, that the author was brother to Pius, Bishop of Rome, and wrote during his Episcopate; that is to say, about the middle of the second century. I have said (p. 52) that I myself believe that statement to be erroneous; but before discussing this point, it will be convenient to say something on some preliminary questions about which there is less room for dispute. If you consider these questions in order, you will be able to judge how far you can travel in my company.¹⁹

(1) Did the author wish his readers to believe that he had actually seen the visions, and received the revelations which he relates? Donaldson (Apostolic Fathers, p. 326) thinks that if Hermas fancied he saw the visions he must have been silly, and if he tried to make other people believe he had seen them, he must have been an impostor. He prefers to think he was neither one nor other; and therefore he looks on the book as belonging to the same class as Bunyan's Pilgrim's Progress, in which edifying lessons are conveyed through the medium of allegorical fiction, which no one is supposed to take as a record of actual facts. It is to me amazing that anyone with ordinary powers of literary perception could read the book of Hermas, and doubt that the author, impostor or not, intended his readers to take him seriously. The judgment I have quoted illustrates what I said (pp. 328, 329), that a man incapacitates himself for historical criticism, if he so takes up the modem attitude of mind towards the supernatural, as not only to disbelieve in it himself, but to be unable to conceive that men in former times felt differently. A man might now publish an edifying fiction in the form of a vision, and without taking any special precaution feel sure that his readers would not imagine he wanted them to take it as real. But in the second century a writer was bound to calculate on a different state of feeling on the part of his readers. And, in point of fact, the Shepherd* was for a time very generally accepted as a record of real revelations. And no critic of early times, whether he accepted the book or not, dreamed that its author wished to convey any other impression.

(2) What, then, are we to think of what Hermas, when relating the circumstances of his visions, tells about himself and his family? If the story be fiction and allegory, we have no right to suppose any of these details to be more real than the angels and towers which he sees in his visions. Nor are we even warranted in assuming that the name Hermas, ascribed to the recipient of the revelations, is that of the author himself. But both the story itself, and the manner of telling it, prove that this is no work of fiction. The author of such a work would strive to give some intelligible account of the hero of his narrative; but here Hermas, as if writing to people who knew him, gives no direct account of himself, and his story has to be deduced by piecing together several incidental notices. What we gather from them is, that Hermas had been brought to Rome as a slave; that Rhoda, the lady to whom he had been sold, set him free, and loaded him with many benefits; that he had acquired some property, and been engaged in trade, which he owns he did not always carry on honestly; that he married a not very handsome wife, who unfortunately was not able to govern her tongue; that he had other trouble with his children, who in time of persecution denied the faith, and betrayed their parents; that he thus lost house and property, but remained steadfast in the faith, and supported himself by agricultural labour. Some have imagined that the * Shepherd was a romance written in the middle of the second century, but intended to have as its hero the Hermas mentioned a hundred years before in the Epistle to the Romans. But it is not credible that the author of a romance would invent for his hero such a history as I have described, representing him not even as a clergyman but a layman, an elderly married man, with an ill-conditioned wife and children. I have dwelt at length upon this point because I am persuaded that the key to all sound criticism on the Shepherd is to understand thoroughly that the Hermas who tells the story is no fictitious character, but a real person, who published his visions for the edification of his contemporaries.

(3) But did he invent these visions, or did he himself believe in them? I have no hesitation in saying that he did believe in them. It is not merely that the whole book impresses me with belief in the narrator's good faith in this respect; but the stories themselves, when examined, show every mark of being, not arbitrary inventions, but attempts to record the imaginations of a dream. I take, for example, the first vision. Hermas relates that he had one day seen his former mistress, Rhoda, bathing in the Tiber, and had assisted her out of the water. And, admiring her beauty, he thought what happiness it were for him had he a wife like her in form and in disposition. Further than this his thought did not go. But soon after he had a vision. He fell asleep, and in his dream he was for a long time walking and struggling on ground so rugged and broken that it was impossible to pass. At length he succeeded in crossing the water by which his path had been washed away, and coming into smooth ground, knelt down to confess his sins to God. Then the heavens were opened, and he saw Rhoda saluting him from the sky. On asking her what she did there, she told him that she had been taken up to accuse him before the Lord, who was angry with him for having sinned against her. He asks her, How? Had he ever spoken a lewd word to her? Had he not always treated her with honour and respect? She owns it, but accuses him of having entertained an evil thought, and tells him of the sin of evil thoughts, and their punishment. Then the heavens were closed, and he was left shuddering with fear, not knowing how he could escape the judgment of God if such a thought as his were marked as sin. Then he sees a venerable lady sitting in a great white chair, with a book in her hands. She asks why he who was usually so cheerful is now so sad. On his telling her, she owns what a sin any impure thought would be in one so single-minded, and so innocent as he; but she assures him it is not for this God is angry with him, but because of the sins of his children, whom he, through false indulgence, had allowed to corrupt themselves; but to whom repentance was still open, if he would warn them. Then she reads to him out of her book: of the greater part he can remember nothing, save that it was severe and menacing; but he remembers the last sentence, which was mild and consoling. She leaves him with the words, Play the man, Hermas.

Now, if we take this story as allegorical fiction, it is impossible to assign a meaning to it. There is not a word more about Rhoda through the whole book. Why has she been introduced? What is she intended to represent? Why should Hermas be first told that God was angry with him on one account, and then be told that it was really on another account God was angry? On the other hand, the want of logical connexion between the parts of the story is explained at once if we take his own word that it was a dream. There is no difficulty in believing that he had seen Rhoda as he tells, and that the thought he had entertained presented itself to him afterwards in his sleep as a sin. It is quite like a dream that Rhoda, as principal figure, should fade out, and be replaced by another; that sensations of physical distress in his sleep should suggest the ideas, first of walking on and on without being able to find an outlet; afterwards of mental distress at words spoken to him; and altogether like a dream, too, that he should imagine himself to have heard a long dis course, yet be able to tell nothing of it but the words heard just before awakening. It therefore seems to me quite false criticism to put any other interpretation on the story told by Hermas than that his visions commenced in the manner he describes, by his having what we should call a very vivid dream. He was much impressed by it, and when, in the following year, he dreamed again of the lady and her book, he regarded it as a divine communication, and set himself, by fasting and prayer, to obtain new revelations. As might be expected, more visions followed, and he made himself known to his Church as favoured with Divine revelations. I see no reason for doubting the truth of this story, though I naturally think that the visions of Hermas gained a good deal in coherence when he came to write them down. I believe, also, that the last two sections of his work contain records of his waking thoughts, which he regarded as inspired by an angel who, he had persuaded himself, had come permanently to dwell with him. The conclusion, then, at which I arrive is, that the work of Hermas is not to be classed with Bunyan's Pilgrim's Progress, but rather with the revelations of St. Teresa, St. Francesca Romana, St. Gertrude, St. Catherine of Siena, and other literature of the same kind, of which there is such abundance in the Roman Catholic Church.

Are we, then, who do not believe in the revelations of Hermas, to set him down as a crazy person, and to regard those who believed in him as fools? The examples I have just cited may make us hesitate before coming to such a conclusion. St. Teresa, for instance, visionary as she was, did much useful work, and exhibited a large amount of practical good sense. In respect of sobriety, the visions of Hermas contrast very favourably with some of the other literature with which I have compared them. I will not discuss the vision of Col. Gardiner, which was accepted as real by Dr. Doddridge, nor need I remind you how many persons who can by no means be described as fools have thought it worth while to record remarkable dreams, under the belief that supernatural intimations might thus have been given. But if you think that the Church of Rome was in the beginning of the second century too easy in its reception of the revelations of Hermas, I will ask you to bear in mind that the men of that age are not to be scorned because their views as to God's manner of governing His Church were different from what the experience of so many following centuries has taught us. We all believe that in the time of our Lord and His Apostles a great manifestation of the supernatural was made to the world. How long, and to what extent, similar manifestations would present themselves in the ordinary life of the Church, only experience could show. Again, if we are able to give a natural explanation of some mental phenomena which were once thought to indicate supernatural interference, it is no dis grace to men of early times that they were not acquainted with modern philosophy. Even in the Church of Rome, though we may think it gives credence too lightly to modern miracles, a visionary would now receive from her spiritual guides instruction as to the possibility of deception, and as to the need of caution, for which, in the second century, no necessity might be felt.

(4) I come, then, to the question, Did Hermas see his visions in the Episcopate of Clement? He himself plainly intimates that he did. For he states that in his vision he received the following instructions: You shall write two books, and send one to Clement and one to Grapte. And Grapte shall admonish the widows and orphans, and Clement shall send it to foreign cities, for to him that office has been committed. And you shall relate it to the presbyters of the Church. The natural inference from this passage is, that at the time of the vision Grapte was what we may describe as chief deaconess of the Roman Church, and that Clement was the organ by which it communicated with foreign Churches. And we have every reason to think that he was so described on account of the celebrity gained not long before by his letter sent to a distant Church. Different ways have been devised of escaping this inference. I really don t know whether we are to count Origen as rejecting the obvious meaning of the passage, though he does manage to find an allegory in it. He treats (De Princip. iv. 1 1 ) of three modes of interpreting Scripture, corresponding to the tripartite nature of man body, soul, and spirit. And he imagines that he finds them indicated in this passage, Grapte, who instructs those of lowest spiritual discernment, being the literal interpretation, and Clement and Hermas himself representing the two higher methods of interpretation. A solution more acceptable to modern habits of thought is that a real Clement is intended, only not the Clement who wrote the Epistle to the Corinthians. But it must be pronounced extremely improbable that within a comparatively few years of the writing of that letter there should be another Clement, whose function it also was to communicate on behalf of the Church of Rome with foreign Churches, but who has left on ecclesiastical history no trace of his existence.²⁰ A third solution is that Hermas, no doubt, wished his readers to believe that he saw his visions in the Episcopate of the well-known Clement; but he was telling a lie: he really wrote forty or fifty years later. But we cannot adopt this solution unless we abandon the results we have already obtained. If the work is a mere fiction, the imaginary hero may have lived under Clement, and the real author when you please; and his name may or may not have been Hermas. But if he was a man who told his contemporaries of visions, real or pretended, which he claimed to have seen himself, it would be absurd of him to destroy his chance of being believed, by asserting that he saw the vision at a time when it was notorious that he had either not been born, or could have been only a child. It is to be remembered that the vision represents him to have been then an elderly married man, with a grown-up family. I must add, that Hermas had no motive whatever for antedating his work. His prophecy announced tribulation close at hand, and only a short intervening period for repentance. It would be absolutely contrary to his interest to pretend that the prophecy had been delivered forty or fifty years previously. All his readers would then know that the prediction had failed, for nothing had come of it. And the promise of forgiveness, which excluded all those baptized after the date of the prophecy, would not be applicable at all to the generation to which the book was offered. I therefore find it impossible to resist the evidence afforded by this passage, that Hermas must have attained to middle life before the death of Clement. I may claim Bishop Lightfoot as agreeing with me in this result; for he repeatedly speaks of Hermas as a younger contemporary of Clement (Philippians, p. 167; Clement, p. i, &c.).

When this result has been adopted, the main question may be regarded as settled. For the remaining point in dispute concerns not the date of Hermas, but the credit due to the Muratorian writer.

(5) If we admit that the vision was seen in the Episcopate of Clement, can we accept the Muratorian statement that Hermas wrote the Shepherd while his brother, the Bishop Pius, sat in the chair of the Church of the city of Rome? Lightfoot thinks we can; and he suggests modes of reconcilement, which, indeed, I tried for a long time myself before I could persuade myself to abandon the Muratorian statement altogether. Hermas may have been considerably the older of the two brothers: perhaps we may give up half the Muratorian statement, and believe that he was the brother of Pius, but not that it was during his Episcopate he wrote the Shepherd; perhaps if we had the Greek of the Muratorian fragment we might not find that assertion there. Then, again, we have not such certain knowledge of the dates of early Roman Episcopates as to forbid our manipulating them a little. Could we not screw up the date of Pius somewhat, and screw down the date of Clement? Possibly we could bring down the date of the death of Clement as late as no; and perhaps we might bring up the accession of Pius earlier than 139, which Lipsius names as the earliest admissible date. But I abandoned these attempts when I saw that a real reconcilement with the Muratorian writer was in the nature of things impossible. His object was to prove Hermas to be quite a modern personage. How could he be that if he had attained the age of forty before the death of Clement?

Let us inquire, then, if we are bound to reconcile ourselves with this writer. Who was he? Had he any real knowledge of the events of the Episcopate of Pius? Critics confess them selves unable to answer the former question, and the majority of those who accept his statement about Hermas answer the second question in the negative. He describes Pius as sitting in the chair of the Church of the city Rome/ and evidently has no suspicion that the constitution of that Church was different in the days of Pius and in his own. But in Hermas the honour of a chair is not confined to a single person, and the critics of whom I speak imagine that Episcopacy was only then struggling, against much opposition, into existence. If the Muratorian writer knew nothing of such a patent fact as the constitution of the Church in the days of Pius, he cannot be an authority as to the date of publication of a book which must have appeared, if not before, at the very beginning of that episcopate. I have elsewhere²¹ given my reasons for thinking that the Muratorian Fragment is a document not earlier than the episcopate of Zephyrinus, that is to say, the beginning of the third century; and I will now mention my theory as to the discovery that the author of the Shepherd was brother of Pius. This discovery is found also in a note appended to a very ancient catalogue of the bishops of Rome. Many good critics have thought that the earlier part of this catalogue was derived from a list made by Hippolytus of the bishops of Rome down to his time, which formed part²² of his Chronology. My theory, then, is that Hippolytus, in the course of the investigations necessary for framing this list, ascertained that Bishop Pius had a brother named Hermas, and that he then jumped to the conclusion (as he was a man quite capable of doing) that this Hermas was the author of the Shepherd. Whether this theory of mine be true or not, I hold that whatever conclusions as to the date of the Shepherd we draw from a study of the document itself ought not to be laid aside in deference to the authority of a writer concerning whose means of information we really know nothing. If no more be granted than Lightfoot has conceded, the date is quite early in the second century, and the Shepherd therefore deserves the highest attention from the student of Church history. And, if it be read without any prepossession to the contrary, I am persuaded that its contents will be found entirely to correspond with that early date, since it reveals an immaturity of development both in respect of doctrine and of Church organization.

The length of the discussion necessary to establish the date of Hermas precludes me from treating of many interesting questions raised by the contents of the book; and I will only say something as to what we may gather from it as to Church organization. It has been the bane of ecclesiastical history that so many have studied it only in the hope to gain from it some weapon which might be used in modern controversies. It is natural to think that if parity of presbyters had been the Church's original rule, the government of a single head could not have been established without some resistance on the part of those who were dispossessed of their equal authority. It has been hoped to find some exception to the almost total silence of Church history as to such resistance, in the language in which Hermas rebukes the strifes for precedence among Christians. I think I am without prejudice in this matter; for I find it much easier to prove from Scripture that individual Christians are bound to submit to the established order of the Church than to prove that the Church had been bound to develop its organization in one particular way. And for me it has only a speculative interest to enquire what was the process by which the Church arrived at the state of things that we find when Church history first comes into clear light at the end of the second century, at which time we find bishops everywhere, and no memory that there had ever been any other form of Church government. But as far as I can see, the question whether one presbyter had pre-eminence over others was one in which Hermas took no interest, and on which he tells us nothing. He clearly distinguishes him self from the presbyters, and makes no claim to be one of their body. But he has something to tell us about the prophets, the class to which, I have no hesitation in saying, he himself belonged. The Church had then its authorized teachers and rulers; but we learn from Mandat. xi. that there were, besides, prophets, or as we may call them, lay preachers. Such a prophet was permitted to give exhortation in the public meetings for worship.²³ After the intercessory prayer had been made, the angel of the prophetic spirit would fill the man, and he would give exhortation to the people as the Lord willed. It is a mark of the antiquity of our document that it indicates that gifted persons were still permitted, as in 1 Cor. xiv. 26, to speak in the Church. It can readily be imagined that the interference of the rulers of the Church would sometimes be necessary to suppress indiscreet or erroneous teaching. It strikes me as possible that the rebellion in the Church of Corinth, where, even in St. Paul's time, spiritual gifts had been exercised without due regard to order, may have originated in an unsuccessful interference of authority with some leading prophets. It was soon found expedient to confine the work of exhortation to the Church's authorized teachers. When, towards the end of the second century, the Montanists brought prophesying again into prominence, precedents in their favour were neither numerous nor then very recent; and it was found that the inspired authority which these prophets claimed threatened to be subversive of all Church order and fixity of doctrine. Hermas belonged to an age when the exercise of prophetic gifts was not discouraged by the Church authorities; but he is distinctly pre-Montanist. I have already mentioned how repugnant his teaching was to the Montanist Tertullian. Hermas occasionally gives indications of some little jealousy²⁴ of the superior dignity of the presbyters. Thus, in one vision, the Church, who appears to him in the form of a lady, bids him sit down. Nay, he modestly answers, let the presbyters be seated first. Sit down, as I bid you, the lady replies. But his chief anxiety is to guard the office of prophet from being intruded on by unworthy persons. Some, it would appear, claimed to be prophets in the modern sense of the word: persons would visit them, ask them questions about their private affairs, and pay money for their advice; and Hermas states that their predictions would occasionally turn out right. But he urges that the Spirit of God does not speak in answer to questions; that is to say, when man wishes Him to speak, but when He Himself chooses to speak. These false pretenders, so ready to prophesy in a corner, are dumb when they come into the Church assembly. Their whole manner of life must distinguish the true prophet from the false: the one is meek, humble, easily contented; not talkative, ambitious, greedy, luxurious, like the other.

The circulation which the work of Hermas obtained gives us reason to think that his own claims as a prophet were admitted by his Church, and that the record of his visions was sent to foreign Churches as he desired. But I can well believe that there had been some hesitation as to recognizing him, and thus that a little soreness of feeling on his part may have arisen. For, though a pious man, he does not appear to have been a well-instructed one; and some of his doctrinal teaching, which is not accurate when judged by the standard of our day, may well have been thought unsatisfactory by the presbyters of his own. He does not formally quote the scriptures either of Old or New Testament; nor does he make much use of either, his coincidences being closest with the Epistle of St. James. It is very possible that he came from the Jewish section of the Church; but, in his work, there is not a trace, not to say of anti-Paulinism, but even of Judaism. In his teaching the Jewish nation has no special prerogative; and even the twelve tribes are only the various nations which make up the Christian Church.

Hernias and Theodotion. Something, however, must be said as to the use made by Hernias of one Old Testament passage; because it has been imagined to afford an argument subversive of the conclusions I have arrived at as to the early date of the work. In the visions of Hermas (iv. ii. 4) he sees a terrible wild beast, from which he is delivered by the protection of the angel who is over the beasts, whose name is Thegri. This Thegri, of whom no one else makes mention, had been a puzzle to commentators until not long since, when the solution was obtained by Mr. Rendel Harris (Johns Hopkins University Circulars, iii. 75). He compares the words in Hermas, ὁ κύριος ἀπέστειλεν τὸν ἄγγελον αὗτοῦ, τὸν ἐπὶ τῶνθηρίων ὄντα, οὗ τὸ ὄνομά ἔστι Θεγρί, καὶ ἐνέφραξεν τὸ στόμα αῧτοῦ ἴνα μή σε λυμάνῃ, with the words of Daniel vi. 22, ὁ θεός μου ἐπέστειλε τὸν ἄγγελον αὔτοῧ, καὶ ἐνέφραξε τὰ στόματα. τῶν λεόντων, καὶ οὐκ ἐλυμήναντό με, , when the use of Daniel by Hermas is seen beyond mistake. But, in the original, the verb corresponding to ἐνέφραξε is סְגַר; and it becomes apparent that we must correct Θεγρί into Οεγρί, and understand the angel who stops the mouths of the beasts.

This remark by Mr. Harris led to a further remark by Dr. Hort. He pointed out (Johns Hopkins* University Circulars, iv. 23) that the strong coincidence between Hermas and the Book of Daniel only exists when Theodotion's version of the latter book is used. The corresponding verse in the LXX. merely has σέσωκέ με ὁ θεὸς σῖπό τῶν λεόντων. In another place, indeed, it has ὁ θεὸς ἀπέκλεισε τᾶ, στόματα τῶν λεόντων; but it neither has ἐνέφραξεν, nor does it use the verb λυμαίνομαι. It follows that Hermas used, not the LXX. version of Daniel, but that of Theodotion; and, therefore, that we must take it as a fixed point in our discussions about the date of Hermas, that he is later than Theodotion; and Theodotion is commonly believed to have made his version in the latter half of the second century.

Now, let me say in the outset, that conclusions drawn from the study of the character of an entire hook are not to be lightly displaced by an argument founded on a single passage. Thus, when treating of the genuineness of 1 Thessalonians, I did not think it worth while to discuss the ingenious little argument which Holsten (see note p. 396) founded on ch. i. 3. In the present case we have in our hands the whole Book of Hernias, containing many notes of time; but we have no trustworthy information as to the date of Theodotion's version, and (what is of more importance) we have no sufficient information what other Greek versions there may have been antecedent to his. We are, therefore, on much firmer ground if we use Hernias to throw light on the history of Greek translations of the Book of Daniel than vice versa. Obviously, we cannot infer from coincidence in a single verse that Hernias was later than Theodotion, if it is possible that in that verse Theodotion himself was but following the lines of an older translator. And that (not to mention Aquila's version, concerning whose rendering of this verse we have no information) there was, in point of fact, such an older translation, has been made almost certain by investigations, on which Dr. Gwynn at first entered for my assistance in dealing with the present question, and which he afterwards carried on on his own account, and the results of which he has published in his articles SYMMACHUS and THEODOTION, in Smith and Wace's Dictionary of Christian Biography.

One preliminary consideration may be mentioned, which may lead us to suspect that there must be some flaw in this argument for the later date of Hernias. The argument proves a little too much; it proves that the Epistle to the Hebrews was also written late in the second century. When the writer of that Epistle uses the phrase (xi. 33), stopped the mouths of lions, we can scarcely doubt that he had Dan. vi. 22 in his mind. We may also take it as certain that he used a Greek, not a Hebrew, Bible. But, if it was the Septuagint version of Daniel that he used, how came he to stumble on the word ἔφραξαν instead of the ἀέκλεισε of the LXX.?

The knowledge which the Christian Church has possessed of Greek translations of the Bible was principally, if not exclusively, derived from Origen's great work the Tetrapla. In the first column of that work he published the version of Aquila, noted for its slavish literalness and ruthless sacrifice of Greek to Hebrew idioms; in the second column, the version of Symmachus, marked by greater purity of Greek; in the third column, the Septuagint; in the fourth, the version of Theodotion, who is said to have been less an independent translator than a reviser of former translations. These were not the only translations which had been made before the time of Origen; for he recovered and published fragments of two or three other versions; but these alone had reached him unmutilated. Of these four, the Septuagint alone is regarded as pre-Christian. Aquila's, which is accounted the oldest of the others, is said to have been characterized by an animus hostile to Christianity, and to have been intended to deprive the Christians of the use of certain O. T. texts on which they had founded arguments. The Septuagint was the Greek version which was used in the Christian Church, and was regarded as inspired by many of the Fathers who accepted a miraculous account of its origin. But there was one re markable exception, the Book of Daniel. St. Jerome states repeatedly that the Christian Church used, not the Septuagint translation of the Book of Daniel, but that of Theodotion. For example, in the preface to his translation of the Book of Daniel, he says: Danielem Prophetam juxta LXX. interpretes, Domini Salvatoris Ecclesiae non legunt, utentes Theodotionis editione; et hoc cur accident, nescio. Sive quia sermo Chaldaicus est, et quibusdam proprietatibus a nostro eloquio discrepat, noluerunt LXX. interpretes easdem linguae lineas in translatione servare; sive sub nomine eorum ab alio, nescio quo, non satis Chaldaeam linguam sciente, editus est liber; sive aliud quid causae extiterit ignorans, hoc unum affirmare possum, quod multum a veritate discordet et recto judicio repudiatus sit (see also the Preface to the Commentary on the Book of Daniel, the Prologue to Joshua, and Apol. cont. Ruf. ii. 33). Thus it appears that Jerome, who was acquainted with the Tetrapla of Origen, took notice that the version of the Book of Daniel in use in the Church of his day was that given in the Tetrapla, not in the Septuagint column, but in the column which presented the version of Thedotion. Jerome is a perfectly competent witness to this matter of fact, though he professes himself unable to offer any but conjectural explanations of it, and though we are unable to accept the explanations which he does give. It would appear that Origen said nothing to throw light on it; though Jerome quotes him as having, at least on one occasion, given, by his example, his countenance to the desertion of the Septuagint forTheodotion: Judicio magistrorum ecclesiae editio eorum (LXX.) in hoc volumine repudiata est, et Theodotionis vulgo legitur; quae et Hebraeo et ceteris translatoribus congruit, unde et Origenes in nono Stromatum volumine asserit se quae sequuntur ab hoc loco in Propheta Daniele, non juxta LXX., qui multum ab Hebraica veritate discordant, sed juxta Theo dotionis editionem disserere (in Dan. iv. 5).

It is, accordingly, Theodotion's version of Daniel which is ordinarily found in Greek Bibles; but the version which stood in the Septuagint column of Origen's Tetrapla has been recovered from a single MS., preserved in the Chigi Library, and was printed at Rome in 1772. It will be found appended to Tischendorf's second and subsequent editions of the Septuagint. An extant Syriac version, and the citations of Jerome, fully establish its claim to be Origen's Septuagint.²⁵ The Roman edition contains a comparison of the variations between the two versions, and a comparison will also be found in the Appendix to Pusey's Daniel the Prophet, p. 606.

Now, to speak first of the date of Theodotion's version, Epiphanius, the earliest writer who gives a date, places it so late, that if Hermas used it, so far from living early in the second century, he could not even have lived in the Episcopate of Pius. In the passage referred to (De Menss. et Pondd., 17), which treats of Greek translations of the Bible, whatever may have been the errors for which Epiphanius is himself responsible, they have been so largely added to by his transcribers, that his Greek text, as printed by Petavius, exhibits a really stupendous mass of blunders. Dr. Gwynn, however, found on consulting, at the British Museum, a Syriac translation, bearing date before A.D. 660,²⁶ that the worst of these blunders can be cleared away; and of those that remain we may charitably believe that some had arisen through negligence of transcribers before the Syriac translation was made. It turns out that Epiphanius means to say that the translation of Symmachus was made in the reign of Marcus Aurelius; and that the translation of Theodotion was made in the following reign, that of Commodus.²⁷

Epiphanius, however, is a writer whose unsupported statements must be received with great caution (see p. 168). I need not inquire how many of his blunders arose from erroneous information, how many from a habit of supplying by invention the defects of his information. In the present case, he is peculiarly untrustworthy, being, on several points, contradicted by older and better authorities. He makes Symmachus an apostate from Samaritanism to Judaism; whereas he really was an Ebionite, as we learn from Eusebius (vi. 17), who had met with a work of his in defence of that heresy. Again, he tells that Theodotion was a native of Pontus, and had been a disciple of Marcion until he became a proselyte to Judaism, when he learned the Hebrew language. But we learn from Irenaeus that Theodotion was really an Ephesian; and we can have little doubt that Epiphanius has mixed up Theodotion with another translator of the Old Testament, Aquila, who was a native of Pontus, and of whom also the story is told that he had been a Christian before he became a proselyte to Judaism. And it would seem to be for no better reason than because he has placed Theodotion at Pontus, that Epiphanius makes him a disciple of the great Pontic heresiarch, Marcion. With respect to his dates, he has certainly placed Theodotion too late in naming the reign of Commodus (180-192). For Irenaeus, who wrote in the beginning of that reign, speaks (iii. 21 ) of the versions of Aquila and Theodotion, and as we shall presently see, his use of the latter translation is such as to show that it could not then have been recent. On the otter hand, Epiphanius has placed Symmachus too early; for Irenaeus does not mention him; and so it is probable that he, and not Theodotion, was the latest of the three translators just lamed. Symmachus was but an older contemporary of Origin, both having had personal acquaintance with the same lady, Juliana (Euseb., as above). Epiphanius appears to have jumped to the conclusion that Symmachus was antecedent to Theodotion, from the fact that, in Origen's columns, the versions stood in the order, Aquila, Symmachus, LXX., TheodDtion, which Origen certainly did not intend as a chronological arrangement. We must, therefore, dismiss Epiphanius's whole account of Greek translations, as being absolutely without historical value. It may not be all pure invention; but we have no means of disentangling the grains of truth it may possibly contain. When we have rejected the testimony of Epiphanius, we are left without any precise information as to the date of Theodotion; but I have no wish to dispute the common opinion that he lived in the second century, because the question with which we are really concerned is whether he did more than revise a previous translation different from the Chigi Septuagint.

Though it is only within very wide limits we can tell when Theodotion lived, we can assign a later limit to the time when his version of the Book of Daniel came into use in the Christian Church. Its use was not due, as some supposed, to the influence of Origen, but is to be found in the previous century. Overbeck has carefully examined (Quaest. Hippol. Specimen, p. 105) the quotations from Daniel made by Irenaeus in his great work on heresies, with the result of finding that Irenaeus habitually uses the version of Theodotion, not that of the LXX. Since we know the greater part of Irenaeus only through the medium of a Latin translation, it might be objected that the quotations only inform us as to the version in use in the time of the translator, and not as to that used by Irenaeus himself. Overbeck, therefore, has pointed out three passages in particular where the argument of Irenaeus turns on words peculiar to Theodotion's version. These are the quotations of Dan. xii. 7, in v. xxvi. 1; of Dan. ii. 44, in v. xx. 1, and v. xxvi. 2. In a citation of Dan. xii. 9, 10, which Irenaeus (i. xvi.) reports as made by the Marcosians, there is a conflation of the two versions. In accepting Overbeck's result, we must guard ourselves by leaving the possibility open that what Irenaeus used was not Theodotion's translation, but an older version closely followed by Theodotion. And when we speak of a conflation, we must always bear in mind the possibility that the so-called conflation may in truth be the earliest document, which may have been partially followed by two independent subsequent writers.

Overbeck has also studied the citations in the work of Hippolytus on Antichrist, and finds, as might be expected from the fact that Hippolytus was a hearer of Irenaeus, that le also used the version of Theodotion. This result is confirmed by Bardenhewer's study of the remains of the work of Hippolytus on Daniel, his report being that Hippolytus not only used the version of Theodotion, but seems ignorant of any other, and that his interpretation sometimes directly contradicts the Septuagint version. Overbeck arrives further at the conclusion that Clement of Alexandria used Theodotion's version (see the passages from Dan. ix., quoted by Clement, that are given by Archbishop Ussher in his Syntagma de LXX. interprett. Versione).

On the other hand, Justin Martyr (Trypho 31) gives a long quotation from Dan. vii., in which the agreements with the Chigi version are so numerous as to preclude the explanation that they result from casual coincidence; and I myself hastily concluded at first that Justin used no other version. But a more careful examination shows that Justin's text exhibits also a number of divergences from the Chigi version, and that in many, though not in all, of these it agrees with Theodotion's. This was observed by Wetstein (Prolegg. in N. T., p. 64, edit. 1730), who, anticipating Dr. Hort in the use of the principle that coincidence with Theodotion proves a writer to be later than Theodotion, drew the inference that the Trypho could not be the work of Justin. Stroth (ap. Eichhorn, Repert. ii. 75), accepting the same principle, inferred that Theodotion must have been earlier than Justin. But Credner (Beiträge, ii. 261-272) gave what I take to be the true explanation, viz. that there must have been an older translation of which both Justin and Theodotion made use.

The citations by Tertullian prove that the so-called LXX. version was accepted as such in Africa at the time that the early Latin translation there used was made. In one work, ascribed to Tertullian, the treatise Adv. Judaeos, Theodotion's version is used. A single example will suffice as illustration. The words (Dan. x. n) translated in our version, O Daniel, a man greatly beloved, are rendered in the LXX. Δανιήλ, ἄνθρωπος ἐλεεινὸς εἶ; but by Theodotion, ἀνὴρ ἐπιθυμιῶν. Now in De Jejun. 9, the passage is quoted in the form, Daniel, homo es miserabilis; but in Adv. Judceos 9, Vir desideriorum tu es. The difference here pointed out goes to confirm Meander's suspicions that the section in which these citations occur is not genuine.²⁸ But the treatise against the Jews, if written by Tertullian, must have been one of his latest works, and full forty years later than the treatise of Irenaeus. It might seem more likely than not that in that interval of time Theodotion's Daniel, which was habitually used by Irenaeus, would have been made by translation accessible to Latin-speaking Christians. Cyprian shows acquaintance with both versions, using, for instance, the LXX. form of Dan. ii. 35, Test. ii. 17; but ordinarily Theodotion: see, for example, Dan. xii. 4, in Test. i. 4.

In view of the facts which have been stated, I find it impossible to accept the received opinion, founded on the authority of the passage in Jerome already quoted, that the Christian Churches up to the middle of the second century used the LXX. version of the Book of Daniel, and afterwards rejected it and replaced it by Theodotion's. St. Jerome, it will be observed, does not profess to have any historical tradition of such a rejection, but merely attests the fact that in his time Theodotion's version was in universal use. But when could such a rejection have taken place, and how could it take place both universally and silently? It must have taken place before the time of Irenaeus, who, as we have said, used a version substantially the same as Theodotion's. I have rejected Epiphanius's statement that Theodotion and Irenaeus both worked in the same emperor's short reign; but unless current opinion as to the date of Theodotion's version be widely mistaken, it must have been quite a modern one in the days of Irenaeus. And it was the work not of a Christian, but of a Jewish proselyte. Now Irenaeus (in. xxi.) believed in the divine inspiration of the seventy interpreters; and in the chapter to which I refer his object is to establish that, in comparison with their work, the versions of Aquila and Theodotion have no authority deserving of regard. Is it then credible that he should, with out a word of explanation, sweep away an entire book of the Bible of these venerated translators, and replace it by the work of an enemy of the Church? Is it not strange, too, that the upstart version should meet as much acceptance in Alexandria as in Gaul? And, again, is it not strange that it should be Theodotion, who of all the ancient interpreters followed most closely the lines of the LXX., and is supposed to have been least acquainted with Hebrew or Chaldee, who should have cast the LXX. completely aside, and made a totally independent translation? I am therefore disposed to believe not only that Theodotion followed the lines of an older version,²⁹ and that this was the one used by Irenaeus; but also that this older version was what Irenaeus recognized as the Septuagint. In fact, our common use of the phrase the Septuagint attributes to that work greater unity than it really possesses. Critics are now agreed that the different books included in it were not all translated by the same hands or at the same time; so that it is really not a single version, but a collection of different versions. If a purchaser now asks for a copy of the Septuagint, the book that goes by that name, which the bookseller will offer him, will contain, not the Chigi version of Daniel, but Theodotion's version. May it not be the case that Irenaeus and Clement had no intention of superseding the Septuagint, but only that the collection to which they gave the name of Septuagint, instead of the Chigi Daniel (which was accepted as part of the Septuagint in Palestine, where Justin Martyr lived and where Origen made his Hexapla), contained a different version; probably not Theodotion s, but the version which was the basis of Theodotion's revision? If this older version was in substantial agreement with Theodotion s, the substitution of the latter version in Church use might easily take place silently.

At all events, an examination of the Chigi Daniel will make it appear intensely improbable that this could have been the only version through which the Book of Daniel was known to Greek-speaking Jews until the second century after Christ. For this version is not so much a translation as a free reproduction of its original, bearing to Theodotion's version the same relation that the Apocryphal First Book of Esdras bears to the corresponding portions of the Canonical Scriptures. Dr. Gwynn's conjecture seems to me well worthy of consideration, that the Apocryphal Esdras and the Chigi Daniel may have had the same author. There is one remark able coincidence between them: ἀπηρείσατο αὐτὰ ἐν τῷ εἰδωλείῳ αὐτοῦ (1 Esdras ii. 10; Dan. i. 2). And the two works resemble each other, not merely in continual arbitrary changes from the original, but in both containing ornamental additions. As the Greek Daniel adds to the Chaldee the stories of Susanna and of Bel and the Dragon, so the Greek Esdras adds the story of the three young men at the Court of King Darius. The latter even contains a hymn after the pattern of the Song of the Three Children, though on a much smaller scale. And, though the Book of Esdras had not the good fortune to be admitted into the Canon of the Council of Trent, no part of the deutero-canonical books has received more extensive Patristic recognition than the story just cited. The Apocryphal Esdras may very possibly be an older translation than the Canonical Ezra; for the latter is a separate book from that of Chronicles; but to all appearance they had formed one book when the translation of the Apocryphal Book was made; and that this was the original form of the Hebrew may be gathered from the identity of the last verse of Chronicles with the first verse of Ezra. This difference of form of the two Greek books prevented them from being taken as different translations of the same book; and so both passed as distinct books into the Greek Bible under the names of First and Second Esdras. But, if the range of contents of the two books had been the same, it might well have happened that the Apocryphal Esdras might have been placed by Origen in his Septuagint column, and the Canoni cal Esdras in the Theodotion column; and then we should have a parallel to what has happened in the case of the two versions of Daniel.

I have just said that it is more probable than not that, long before the second century after Christ, the Chisian version should have had to encounter the rivalry of a more faithful translation; and it might perhaps be supposed that the facts already brought forward could be explained by pushing back the date of Theodotion's translation to the early part of the second century. But a table, drawn up by Dr. Gwynn, of the New Testament citations of Daniel with the corresponding renderings in Theodotion, and in the so-called Septuagint, proves decisively the existence of a version different from the Chisian, at an earlier date than it is possible to imagine Theodotion to have lived. Instead of this table exhibiting an exclusive use of the Chisian version, it is really surprising how little evidence it affords that that version was even known to the N. T. writers, though it must certainly have been in existence long before their time. I have already referred to the Epistle to the Hebrews. The Apocalypse is the N. T. book which makes most use of the Book of Daniel. In that book the result of the comparison is, that there are several passages in which St. John does not use the LXX., and does approach nearer to Theodotion; and that there is no thing decisive the other way. So that I actually find in the Apocalypse no clear evidence that St. John had ever seen the so-called LXX. version. The following are some of the pas sages in question:

(1) Rev ix. 2O: τὰ εἴδωλα τὰ χρυσᾶ καὶ τὰ ἀργυρᾶ καὶ τὰ χιιλκᾶ καὶ τὰ λίθινα καὶ τὸ. ξύλινα ἃ οὔτε βλέπειν δύνανται οὔτε ἀκούειν οὔτε περιπατεῖν. There is not a word of this in the LXX.; but Theodotion has, Dan. v. 23, τοὺς θεοὺς τοὺς χρυσοῦς καὶ ἀργυροῦς καὶ χαλκοῦς καὶ σιδηροῦς καὶ ξυλίνους καὶ λιθίνους, οῗ οὔ βλέπουσι καὶ οἳ οὐκ ἀκούουσι.

(2) Rev. x. 5: ὤμοσεν ἐν τῷ ζῶντι. So Theod. (Dan.xii. 7); but LXX., ὤμοσε τὸν ζῶντα.

(3) Rev. xii. 7: Μιχαὴλ . . . τοῦ πολεμῆσαι. Theod. has also τοῦ πολεμῆσαι (Dan. X. 20); but LXX., διαμάχεσθαι without τοῦ.

(4) Rev. xiii. 7: πόλεμον μετὰ τῶν ἁγίων. So Theod. (Dan. vii. 21); but LXX., πρὸς τοὺς ἁγίους.

(5) Rev. xix. 6: φωνὴ ὄχλου. So Theod. (Dan. x. 6); but φωνὴ θορύβου.

(6) Rev. xx. 4, and Dan. vii. 9. Apoc. and Theod. have κρῖμα: LXX., κρίσις.

(7) Rev. xx. 11: τόπος οὐκ εὑρέθη αὐτοῖς. So Theod. (Dan. ii. 35); but LXX., ὥστε μηδὲν καταλειφθῆναι ἐξ αὐτῶν.

If the first or the last of these examples had been found in Hermas, instead of in the Apocalypse, it would certainly have been regarded as affording positive proof that Hermas used Theodotion. In the present case it may be said that St. John was not under the necessity of using any version, and could have translated for himself from the Chaldee. And so, no doubt, he could. And yet, I think nothing but a strong preconceived opinion that St. John could have used no other version than the Chisian would prevent the conclusion from being drawn that he actually does use a different version. The author of the Apocalypse did not write Greek with such facility that he should scorn to use the help of a Greek translation; and in fact, in the case of other books of Scripture, he shows himself acquainted with the Greek Bible.

If no other version than the Chisian was accessible to St. John, we need not be surprised at his rejecting it and preferring to render for himself, because such a course would certainly be adopted by any Jew who was able to read the original, and who at all valued faithfulness of translation. But is it intrinsically probable that for centuries every Jew competent to ascertain the fact kept to himself his knowledge of the unfaithfulness of the current version; and that none had the charity to make a better version for the use of his Greek-speaking brethren? On the other hand, is it very improbable that such a version, if made, should now only live for us in its successors, as Tyndale's translation lives for us in the Authorized English version?

I think that some of the coincidences noted above, between St. John and Theodotion, especially the τοῦ πολεμῆσαι of No. (3) are more than accidental; but that St. John used a translation of some kind appears more clearly from the very numerous passages where Theodotion and the Chisian agree, and St. John agrees with both a thing not likely to happen so often if he was translating independently. But if St. John used a translation, that translation was not the Chisian, with which he gives no clear sign of agreement. I find instances which may induce us to think that the version employed by St. John was not identical with Theodotion's, but scarcely anything to show that it was the Septuagint. I only notice two cases where, on a comparison of the Apocalypse with the so-called LXX. and Theodotion, the advantage seems to be on the side of the LXX. These passages are:

(1) Rev. i. 14: ἢ κεφαλὴ αὐτοῦ καὶ αἱ τρίχες λευκαὶ ὣς ἔριον λευκόν, ὡς χιών, καὶ οἱ ὀφθαλμοὶ αὐτοῦ ὡς φλὸξ πυρὸς καὶ οἶ πόδες αὐτοῦ ὅμοιοι χαλκολιβάνῳ, Dan. vii. 9 (LXX.), ἔχων περιβολὴν ὧσεὶ χιόνα καὶ τὸ τρίχωμα τῆς κεφαλῆς αὗτοῧ ὥσ εἲ. ἔριονλευκὸν καθαρόν· (Theod.),τὸ ἔνδυμα αὖτοῦ λευκὸν ὡσεὶ χιών, καὶὴ θρὶξ τῆς κεφαλῆς αὐτοῦ ὡσεὶ ἔριον καθαρόν, Dan. X. 6 (LXX.), οἱ ὀφθαλμοὶ αὐτοῦ ὡσεὶ λαμπάδες πυρὸς . . . καὶ οἳ πόδες ὡσεὶ χαλκὸς ἐξαστράπτων (Theod.), οἶ ὀφθαλμοὶ αὐτοῦ ὡσεὶ λαμπάδες πυρὸς . . . καὶ τὰ σκέλη ὧς ὅρασις χαλκοῦ στίλβοντος.

(2) Rev. xix. 16, βασιλεὺς βασιλέων καὶ κύριος κυρίων, So LXX. (Dan. iv. 31), Θεὸς τῶν θεῶν καὶ κύριος τῶν κυρίων καὶ βασιλεὺς τῶν βασιλέων, to which there is nothing corresponding in Chaldee or Theodotion. The former example proves, if proof were necessary, that St. John was not dependent on Theodotion's version; but does not prove that he used the LXX. I do not know that any stronger proof of that can be given than whatever the latter example may be thought to afford.

Dr. Gwynn has also examined the use made of Daniel in other N. T. books, and still with the result that that use can not be accounted for on the supposition that the N. T. writers used only the Septuagint version of Daniel. For example, the words κατασκηνοῦν and ἐν τοῖς κλάδοις, which occur Matt. xiii. 32, are found in Theodotion's version of Dan. iv. 9; but not in the LXX., which instead of κατασκήνοῦν has ἐνόσσευον.

So in Matt. xiii. 43, Dan. xii. 3, the ἐκλάμψουσιν agrees with the λάμψουσιν of Theodotion against the φανοῦσιν of the LXX.; and in Matt. xxiv. 21, Dan. xii. 21, Matthew and Theodotion agree in θλῖψις οἵα οὐ γέγονεν, where LXX. has ἡμέρα θλίψεως οἵα οὐ ἐγενήθη.

In Mark xiv. 62, as also in Rev. i. 7, the Son of Man is spoken of as coming with (μετά) not on (ἐπί) the clouds; in this agreeing with Theodotion's text against the Chisian. A more doubtful case of coincidence is James i. 12, which is closer to Theodotion's version of Dan. xii. 12 than to the LXX.

Again, Clement of Rome (c. 34) quotes Dan. viii. 10: Ten thousand times ten thousand stood before him, and thousand thousands ministered unto him; and for ministered he has Theodotion's word ἐλειτούργουν, not the LXX. ἐθεράπευον.

Further, the Apocryphal Book of Baruch contains several verses taken from Dan. ix.; Baruch i. 15-18, being nearly identical with Dan. ix. 7-10, and Baruch ii. 11-16, with Dan. ix. 15-18. A few critics bring down this book (or the Greek of it, if we are to receive it as a translation from a Hebrew original) as late as the reign of Vespasian, and none brings it later; but the great majority regard it as pre-Christian. Now, on comparing the passages, Baruch is found to be consider ably nearer Theodotion than the LXX. Thus:

The instances adduced not only clearly prove all I want to establish, namely, that coincidences with Theodotion's version do not prove that a document is not as early as the first century; but they seem to point distinctly to the existence at least in that century, and probably much earlier, of a version of the Book of Daniel having closer affinities with Theodotion's than with the LXX.

The passage in Hernias then simply takes its place as one of many proofs of that fact. I have given these proofs at greater length than was at all necessary for my immediate purpose, on account of the interest I felt in Dr. Gwynn's investigations, which throw light on a subject that has been very little studied: that of the history of first-century Greek translations of the Old Testament. Enough has been said to show that if it can be established on other grounds that the Book of Hermas belongs to the early part of the second century, no reason for rejecting that date is afforded by the fact that we find in the book a verse of Daniel quoted in a form for which the Hexaplar Septuagint will not account.

The Teaching of the Twelve Apostles. It would evidently be impossible for me to keep within reasonable limits if I were to attempt to speak of all the remains of early Christian antiquity which present interesting subjects for discussion. I have therefore taken as my guide the list of works whose claims to be included in the public use of the Church Eusebius thought it worth while to take into consideration when making his list of canonical books (H. E. iii. 25). Of the books there mentioned there remains but one which I have not yet noticed. In company with the Epistle of Barnabas, Eusebius names what are called the Teachings of the Apostles (τῶν ἀποστόλων αἱ λεγόμεναι διδαχαί). I have already (see p. 581) referred to the list of canonical books given some years later by Athanasius, in his 39th Festal Epistle; and there you find, excluded from the books of Scripture, but joined with the Shepherd of Hermas, as useful for employment in catechetical instruction, what is called the Teaching of the Apostles (Διδαχὴ καλουμένη τῶν ἀποστόλων): you will observe that the singular number is used. The Διδαχὴ ἀποστόλων is also included in the Stichometry of Nicephorus (see p. 178). It is found there in an appendix giving a list of apocryphal books of the New Testament, viz. the Travels of Peter, of John, of Thomas, the Gospel of Thomas: then follows the Didaché, and then books to which the name apocryphal can only be applied in the sense that they have no claim to possess the authority of Scripture, viz., the Epistles of Clement, of Ignatius, of Polycarp, and the Shepherd. In this list the length of the Διδαχή is given as 200 στίχοι,³⁰ by which we see that it was a short book, since in the same list the Apocalypse of St. John is said to contain 1400 στίχοι.

Until very recently we could only form a vague judgment that the work known to Athanasius and Eusebius must have been the nucleus round which gathered the institutions which form the extant eight books of Apostolic Constitutions. It is now agreed that this work, in its present form, is not earlier than the middle of the fourth century; and in recent times much has been done to trace the history of the growth of the collection. The subject is too wide a one for me to attempt to enter into it; but it is necessary to mention an ancient tract, the foundation of Egyptian Ecclesiastical Law, first published in Greek from a Vienna MS. by Bickell (Geschichte des Kirchenrechts, 1843), but extant also in Coptic, Æthiopic, Syriac, and Arabic. Bickell called it Apostolische Kirchenordnung; and, in order to distinguish it from the Apostolic Constitutions, which, in their present form, are certainly a later work, I shall refer to this under the name of the Church Ordinances. Its title in the Greek MS. is αἶ διαταγαὶ αἱ διαὶ Κλήμεντος καὶ κανόνες ἔκκλησιαστικοὶ τῶν ἁγίων ἀποστόλων.  It may be divided into two parts; in the first each of the Apostles is introduced as giving a piece of moral instruction; in the second part the Apostles in like manner severally give directions about ordinations and other Church rites. I may mention that the number of twelve Apostles is made out in a singular way. Cephas is made an Apostle distinct from Peter: he and Nathanael take the place of James the Less and Matthias. Paul is not mentioned at all. Now, when this tract is compared with the seventh book of the Apostolic Constitutions, the latter is found to begin with a large expansion of the moral instruction contained in the first part of the former; and the conclusion suggests itself that this tract was one of the sources employed by the compiler of the Apostolic Constitutions. Further, this moral instruction begins with what we may regard as a commentary on Jer. xxi. 8, Behold I set before you the way of life and the way of death words which may themselves be connected with Deut. xxx. 15, See, I have set before you this day life and good, death and evil. The Church Ordinances set forth in detail the characteristics of these Two Ways. One sentence of this exposition is quoted by Clement of Alexandria as Scripture (Strom, i. 20, p. 377), whether he got it in the Church Ordinances themselves, or in an earlier document, from which they borrowed, My son, be not a liar; for lying leads to theft.

The use of an earlier document is made probable by our finding elsewhere this teaching about the Two Ways The Epistle of Barnabas consists of two parts. The first part, which contains the doctrinal teaching, is brought formally to a close in ch. 17, and then the writer abruptly says, Let us now pass to another doctrine and teaching (γνῶιν καὶ διδαχήν). And then he proceeds to give the teaching of the Two Ways, presenting numerous coincidences with the corresponding section in the Church Ordinances. Now, a curious fact is, that this second section of Barnabas is not extant in the ancient Latin translation; whence suspicion has arisen as to the genuineness of this portion of the Epistle. But any hesitation as to accepting the testimony of the Greek text is removed by the fact that passages from this section are expressely quoted as from Barnabas by Clement of Alexandria (Strom, ii. 18, p. 471), and by Origen (De Princ. III. ii. 4). And it may be added, as bearing on the question presently to be considered, whether Barnabas was original in this part of his teaching, that Origen, at least, appears to consider him so, quoting him as the authority for the teaching concerning the Two Ways. The probable explanation of the omission of this section by the Latin translator is, that he left it out because the West was already in possession of the teaching concerning the Two Ways in another form. Evidence of the existence of such a form is found in the commentary on the Creed by Rufinus, written towards the end of the fourth century. He gives (cc. 37, 38) a list of canonical and ecclesiastical books, founded on that of Athanasius; but whereas Athanasius couples the Didaché with the Shepherd, Rufinus has in the corresponding place, libellus qui dicitur Pastoris, sive Hermas; qui appellatur Duae vise, vel Judicium Petri. Now, it is to be observed, that whereas Eusebius (iii. 3), enumerating the apocryphal books bearing the name of the Apostle Peter, gives the titles of four works the Acts, the Gospel, the Preaching, and the Revelation of Peter Jerome in his Catalogue adds a fifth, the Judgment of Peter. We cannot but think that the works mentioned by Rufinus and Jerome are the same; and the second title, the Two Ways, leads us to think that it must have contained the same matter as is found in the second part of Barnabas, and in the Church Ordinances, only that instead of this teaching being, as in the latter book, distributed among the Apostles, it was apparently, in the Western book, put into the mouth of Peter.

The facts of which I have given a summary were discussed in an able Paper by a Roman Catholic divine, Krawutzcky, in the Theol. Quartalschrift, 1882, who drew from them the following inferences: that, as early as the second century, the section in Barnabas which treated of the Two Ways was expanded and formed into a separate tract; that it came into Church use, and was the work cited as Scripture by Clement of Alexandria; that, to give greater weight to the teaching, it was put into the mouth of Peter; that this work was made use of by the compiler of the Church Ordinances, who made the alteration of distributing the teaching among the twelve Apostles; that the compiler of the seventh book of the Apostolic Constitutions, without any acquaintance with the Church Ordinances, made independent use of the Two Ways; so that by comparison of the Constitutions and 4 Ordinances, a restoration of the earlier work which furnished a common element to both might be obtained.

Within two years scholars found reason to think that it was quite true that the Constitutions and Ordinances had a common source, but that there was no need of conjectural restoration in order to recover it. I have related (p. 578) the discovery by Bryennius at Constantinople of a complete copy of Clement's Epistles. The same volume contained other Ecclesiastical writings, and in particular a complete Greek text of Barnabas. The attention of th& discoverer seems at first to have been quite absorbed by the use to be made of his volume in restoring the text of previously known documents; and though he published his edition of Clement in 1875, it was not till the close of 1883 that he gave to the world a previously unpublished work contained in the same volume. This bears the heading Teaching of the Twelve Apostles (Διδαχὴ τῶν δώδεκα ἀποστόλων), and commences, Teaching of the Lord by the twelve Apostles to the Gentiles. It then goes on to give the teaching of the Two Ways, which occupies the first half of the tract. Then follows a second part, giving directions first about baptism, then about Eucharistic formulae, then about Church teachers, and in conclusion there is an eschatological passage treating of the Second Coming of our Lord. This work bears every mark of very great antiquity; and it has been commonly accepted as belonging to the be ginning of the second century, if not to the latter part of the first. And it has been generally recognized as the work known to Eusebius and Athanasius, and as the common source of Ordinances and Constitutions. Krawutzcky, however, resists the temptation to regard the Didaché as the fulfilment of his critical anticipations. He maintains that the result of a comparison of the Ordinances and the Didaché is not that the one book borrows from the other, but that both have employed a common source. And he holds that the Didaché displays Ebionite tendencies, and was probably not written before the close of the second century. And it is quite true that there is much in the book that not only a Roman Catholic, as Krawutzcky is, might naturally dislike to accept as orthodox teaching, but with which even a member of our own Church cannot feel satisfied.

I do not count among reasonable causes of offence that the book displays great immaturity of Church organization, but rather accept this as a proof of the great antiquity of the document. In that part which treats of Church teachers the foremost place is given to Apostles and Prophets. But the word Apostle has not the limited meaning to which modern usage restricts it. The Apostles are wandering missionaries or envoys of the Churches. Directions are given as to the respect to be paid to an Apostle, and the entertainment to be afforded him by a Church through which he might pass; but it is assumed that he does not contemplate making a permanent stay. On the contrary, if he demands lodging for more than two nights, or if on leaving he asks from his entertainers a larger supply than will suffice to carry him to his next lodging, he shows that he is no true prophet. Now, the word ἀπότολος was in Jewish use applied to messengers sent by the rulers at Jerusalem with letters to Jewish communities elsewhere;³¹ it is used in the New Testament of envoys or commissioned messengers of the Churches (2 Cor. viii. 23; Phil. ii. 25); but those are called in a special sense Apostles who derived their commission not from men, but from Jesus Christ. Hermas, also (Sim. ix. 15), appears to use the word in a wide sense, representing the building of the Church as effected by forty apostles and teachers, and these as not holding the foremost place in the work. The use of the word, therefore, in the Didaché affords no cause of offence, but attests the antiquity of the document. The chief place in the instruction of the local Church is assigned to the prophets whose utterances were to be received with the respect due to their divine inspiration, and who were entitled to receive from their congregations such dues as the Jews had been wont to render to the high priests. The possibility is contemplated that in the Church there might be no prophet. In that case the first-fruits are to be given to the poor. Mention is also made of teachers, by which I understand persons who gave public instruction in the Church, but who did not speak in the spirit as the prophets did. The place assigned to the prophets corresponds very well with the state of things which I infer from Hermas, but with this notable difference, that in Hermas the prophets appear to be subordinate to the presbyters. Here, on the contrary, the first mention is only of apostles and prophets; then directions are given for Sunday Eucharistic celebration, and then is added elect, therefore,³² to yourselves bishops and deacons. These, we are told, are to be honoured with the prophets and teachers, as fulfilling like ministration. The inference then suggests itself that at the time this document was written the Eucharist was only consecrated by the president of the Church assembly, who held a permanent office, and who, probably, might also be a preacher; but that in the mind of the writer the inspired givers of public instruction held the higher place. No mention is made of the necessity of obedience to any central authority at Jerusalem, Rome, or elsewhere. Whether the state of ecclesiastical organization here indicated agree or not with what we may think likely to have existed in Apostolic times, and whether we accept the author as a witness to the general practice of the Church in his time, or only as to that which prevailed in his own locality, or according to his own notions of fitness, still there is no reason for setting him down as a heretic, and the unlikeness of his account to the constitution which we know became general before the second century was far advanced, may be taken as proof of the writer's antiquity.

I find much more cause of offence in the Eucharistic prayers which are given (cc. 9, 10). In the first place, we are surprised to find information given as to the most sacred mysteries of the religion in a document clearly intended for the instruction of catechumens. It is free to us, no doubt, to suppose that in that early age no reserve was practised; but Athanasius recommended that the book known to him as the Didaché should be employed in catechetical instruction. Would he use it for such a purpose if it revealed what only the faithful know? These Eucharistic prayers themselves contain no mention of our Lord's institution of the rite, and no mention of His Body and Blood. And through the whole document I find no unequivocal proof that the writer really believed in our Lord's Divinity, or that he looked on Him as more than a divinely commissioned teacher. Krawutzcky remarks that the writer is silent as to the doctrines of the Incarnation and Redemption and of the sending of the Holy Ghost. Still, if he was an Ebionite, he belonged to the better sort of them; he is certainly no Elkesaite. He gives directions for the blessing of the Cup; but in the ascetic sect from which the pseudo-Clementines emanated, wine does not seem to have been employed, even in Eucharistic celebration.

In deciding as to the date of the Didaché, a crucial question is the determination of its relation to Barnabas and Hennas. In the case of Barnabas the obligations on the one side or the other are too extensive to admit of dispute. The parallel passages of Barnabas occupy four pages in Bryennius's edition. Bryennius himself entertains no doubt that the Didaché was indebted both to Barnabas and Hernias, and this view is also taken by Hilgenfeld, Harnack, and Krawutzcky. Eut Zahn and other good critics hold the opposite opinion; and they advance arguments which seem to me to prove decisively that in that part of the Didaché which treats of the Two Ways there is no obligation to Barnabas. The precepts in the Didaché are systematically arranged, following the order of the Decalogue, on which they serve as a commentary; in Barnabas they are quite promiscuous. It is not a probable hypothesis that the author of the Didaché went through Barnabas, picking out the moral precepts, and that he succeeded in arranging his excerpts into a symmetrical whole. Yet if I am right in referring Barnabas to the decade A. D. 70-80, if the Didaché was so much older, and had so much authority as to be thought worth pillaging by Barnabas, its claims to be really an Apostolic document deserve serious consideration; and how are we to explain the very limited circulation which this truly Apostolic teaching obtained, so that it has had the very narrowest escape of perishing altogether?

In solving this difficulty I have found the greatest assistance from a study of the Didaché in connexion with the Talmud, by Dr. Taylor.³³ It results from his investigations that the Didaché is an intensely Jewish document, and that its contents are so well accounted for by the use of Jewish sources, that we lose all temptation to imagine that the author had need to resort to Barnabas for guidance. But Dr. Taylor's illustrations did more than bring me to his conclusion that the author of the Didaché had received a Jewish training; they seemed to me to make it probable that the Two Ways is a pre-Christian work: in other words, that the author of the Didach has taken a Jewish manual of instruction for proselytes, and has adapted it for Christian use by additions of his own; in particular by insertions from the Sermon on the Mount. This hypothesis would account for the heading, Teaching of the Twelve Apostles to the Gentiles. It has been remarked by several that there is nothing in the work which suggests that it is intended for exclusively Gentile use; nay, that as I have intimated before, it does not even seem adapted for the use of catechumens, Jews, or Gentiles. But the title would be accounted for if the original of the document were a manual of instruction for Gentile proselytes to Judaism. There was at least sufficient inducement to take this as a working hypothesis, and see how it would bear examination. For there is a test which can be applied to it, namely, to examine whether Barnabas knew the Didaché in its pre sent Christianized form. If he did, Barnabas was so early that it is unreasonable to assume that there was an earlier form. On the other hand, if Barnabas knew, not the Didaché, but the supposed Jewish parent of the Didaché, it is likely that when he adapted it to the use of his Christian disciples, the Jewish element in the work would no doubt remain the same as in the Didaché; but that the additions of specially New Testament teaching would, except for some chance coincidences, be different. Now, when we look at the four pages in Bryennius which contain Barnabas's adaptation of the Two Ways we find that he has not Christianized it at all. There is no use of the Gospels, no mention of Jesus Christ, not a word that might not have been written before our Lord was born. I do not know how it will appear to others, but to my mind it comes with the force of demonstration, Barnabas never saw the Didaché I find it impossible to believe that if he knew that work he would have gone over it, adapting it to his use by carefully erasing every line which contained anything of specially Christian teaching, or which implied a knowledge of oral or written Gospels. Traces of such knowledge may be found in other parts of the Epistle of Barnabas, but not in this section. The supposition that the Didaché had a Jewish original be comes thus something more than a mere hypothesis: it is a conclusion forced on us if we believe that Barnabas did not use the Didaché, and that the Didaché did not use Barnabas. The difference of order in the two documents is at once explained. The author of the Didaché wrote with the Jewish original before him, and systematically followed its order; Barnabas, merely in giving practical exhortation, interwove, as his memory furnished them, precepts from a manual with which he had formerly been familiar.³⁴ And if he did not reproduce very accurately either the language or the order of the document he used, this, as Dr. Taylor has remarked, ought not to surprise anyone who considers how Barnabas deals with the Old Testament.

If we admit that the Didaché is but a Christianized form of an originally Jewish book, the question whether the writer who gave the work its present form knew Barnabas assumes a different aspect. For, besides the section on the Two Ways, common to both books, there is one clear coincidence between the early part of Barnabas and the last chapter of the Didaché, an entirely Christian chapter, which treats of the Second Coming of our Lord. If I am right in supposing that Barnabas did not know the Didaché in its present form, the obligation cannot be on his side. On the other hand, all the marks of superior antiquity that have been found in the Didaché belong to the Jewish element in the book, so that there is no reason for denying an acquaintance with Barnabas on the part of the writer who contributed the Christian element. There is a difficult phrase in this last chapter, which, if we could only be sure that we interpret it rightly, would afford a more direct proof of the dependence of that chapter on Barnabas. It gives as the first of three signs of our Lord's immediate coming, σημεῖον ἐκπετάσεως ἐν οὐρανῷ. I think Archdeacon Edwin Palmer has given the best explanation of this. He refers to the words of Isaiah (lxv. 2), I have stretched forth (ἐξεπέταα) my hands to a disobedient and gainsaying people. Barnabas interprets this of our Lord's stretching forth His hands on the cross; and Justin Martyr (Apol. i. 35; Trypho 197) and several other fathers follow him in giving this mystical meaning to the verb ἐκπετάννυμι. If we could count the author of the Didaché in the number of these followers, his phrase is at once explained as meaning the sign of the cross. If this explanation be right, the relative order of Barnabas and this part of the Didaché is determined. If Barnabas came first, the phrase in the Didaché is explained; but if the Didaché came first, a phrase so obscure would never suggest to Barnabas his interpretation of Isaiah, and without that interpretation we should be at a loss to know how the phrase came to be adopted.

The coincidences between the Didaché and Hermas are much fewer than in the case of Barnabas, and there is even room, for dispute whether there is literary obligation on either side. I myself did not regard the coincidences as accidental; but, believing Hermas to be later than Barnabas, I did not care to give the matter much examination, it being likely enough that the Jewish document which was used by Barnabas would be known also to Hermas, and used by him. Further consideration, however, has led me to give a different explanation of the coincidences in question. The most important are to be found in the first chapter of the Didaché, a chapter in which large use is made of our Lord's Sermon on the Mount. Now, Gebhardt has brought to light a short Latin fragment, containing the commencement of the Teaching of the Apostles; and in this there is no trace of the Sermon on the Mount section. That this is no accidental omission is proved by a passage in Lactantius (Div. Inst. vi. 3), where he delivers the doctrine of the Two Ways in a form agreeing with the Latin fragment in all the points in which that differs from Bryennius's Didaché, so as to leave no doubt that the fragment truly represents the form in which the Teaching circulated in the West.³⁵ Neither is any trace of the Sermon on the Mount section found in writings ascribed to Athanasius which appear to use the Didaché; and it has been already stated that the length set down for the Didaché in a stichometry favours the opinion that the work intended was a shorter one than that published by Bryennius.

While, then, I hold to the opinion that the original nucleus of the Didaché was pre-Christian, I am led to the further conclusion that Bryennius's form is not that even in which the Didaché first appeared as a Christian document. On examining the coincidences with Hermas which are to be found in the Sermon on the Mount section, which there is reason to regard as belonging to the latest part of the document, the priority seems to me to be clearly on the side of Hermas; and therefore I consider that, though the earliest Christian form of the Didaché" is probably earlier than Hermas, the section in question was added by one who had read Hermas. He who added this section may well have added more; and I am inclined to believe that the other places where coincidences with Hermas occur are also not original. The earliest proof of acquaintance with the Didaché in its present form is to be found in the Apostolic Constitutions, compiled after the middle of the fourth century, which exhibit a clear use of Bryennius's Didaché The document which I have called the Church Ordinances is clearly later than Barnabas, but antecedent to Bryennius's form. I do not venture to say whether the compiler used the Didaché in its Jewish or in its early Christian form.

The theory, then, about the Didaché which most commends itself to me is that it had for its original a form used by Jews for the instruction of proselytes: that this form continued to be used in the Palestinian Churches, with some slight additions and alterations, giving it a more Christian aspect; that the document (being intended, not for literary circulation, but for practical use) received additions from time to time; and that when it came to be known outside the Churches of Jewish descent, it circulated first in its shorter, afterwards in a longer, form. I do not believe that the work at any time had extensive influence or circulation. The testimonies exhibiting knowledge of the existence of a book of Apostolical Teaching appear to me to be really few. I do not find, for example, in the extant works of Irenaeus³⁶ or Tertullian, evidence of know ledge of the existence of such a book. Clement of Alexandria might have brought a copy from Palestine to Egypt; but this I take to have been in the shorter form, which alone is heard of until the time of the Apostolic Constitutions.

If the view I have taken be correct, that the Didaché, as we know it, was a work of very limited circulation and influence, which spread but little and slowly outside the purely Jewish section of the Church, it ceases to be of much importance in the history of the Christian Church; and any inferences we draw from it are affected by the uncertainty whether certain portions of the book, as we now have it, belong to the earliest form. But, on the other hand, the book gains in importance when regarded as a contribution to the history of Judaism, exhibiting the religious training which had been received by pious Jews before the Gospel was preached to them. I therefore turn back to examine how much of the Didaché can be supposed to have been based on a previously existing Jewish manual. To that manual we naturally refer the first five chapters containing the Two Ways. The sixth is a short chapter, giving license to the disciple, in matters of food, not to bear the whole yoke if he is not able, but insisting on his at least abstaining from things offered in sacrifice to idols. Nothing forbids us to think that this was a rule of life prescribed by Jews to a proselyte, and the whole chapter may have been found textually in the original manual.

The seventh chapter treats of baptism. The candidate is previously to have been taught all the preceding instructions; then he is to be baptized in the name of Father, Son, and Holy Ghost. The baptism is to take place in preference in running water; if this cannot be had, in standing water; if cold water cannot be had, it may take place in warm water; by which we are apparently to understand that if neither river nor pond were accessible, the baptism might take place in drawn water, such as that of a bath. If water in sufficient quantity could not be had, water might be thrice poured on the head in the name of Father, Son, and Holy Ghost. Both baptizer and baptized were to fast previously, and, if possible, others with them; but in any case the person to be baptized must fast beforehand one day or two. It is evident this chapter has been Christianized; but the original document could hardly have failed to contain in the corresponding place instructions about baptism, which was a ceremony considered essential in the admission of proselytes. The doctrine of the absolute necessity of the preliminary fast receives a curious illustration from the pseudo-Clementines. In the part of that romance (Recog. vii. 36; Horn. xiii. u) which relates the baptism of Clement's mother, Peter directs that she must fast one day previously. She declares that she has eaten nothing for the last two days (a fact to which Peter's wife bears witness), and asks to be baptized at once. Peter smiles, and explains that a fast made without reference to baptism will not count. She must fast all that day; they will all fast with her, and then she can be baptized the next day.

The next chapter in the original in all probability treated of fasting and prayer. The Didaché here directs the disciple to fast twice a-week; but not on Mondays and Thursdays, like the hypocrites, but on Wednesdays and Fridays; and to pray three times a-day; but instead of praying like the hypocrites, to use the Lord's Prayer, which is given with the doxology. It appears to me that the adapter here designedly departed from his original; and that the rules of fasting and the prayers which he calls of the hypocrites, were those which he found in his original, and for which he substitutes purely Christian equivalents. Epiphanius (Haer. 16) speaks of the Monday and Thursday fast as a Pharisaic institution. The author of the Didaché had, no doubt, in his mind our Lord's words, which occur so often in Matt, xxiii., Scribes and Pharisees, hypocrites!

The ninth and tenth chapters of the Didaché are generally understood as referring to the Eucharist. I have already intimated some difficulty as to this view, and the difficulty is increased by the fact that the Eucharist is treated of in a later chapter (14). Why should it be treated of twice? I believe the answer to be, that in the corresponding place of the original Jewish manual the proselyte was taught as the concluding piece of his instruction forms of benediction to be used before and after solemn meals. These forms, I take it, the compiler of the Didaché adapted for Christian use, leaving it free, however, to persons endowed with prophetical gifts to use different forms if they chose. These forms might be used in the Christian Love Feasts; but I do not believe that the Eucharist proper is treated of before the fourteenth chapter. And, in fact, if I am right in my inference from the therefore at the beginning of chap, xv., the Didaché agrees with Justin Martyr in making consecration the office of the president of the assembly, and there could be no reason why formulae for the purpose should be taught to the ordinary disciple. It is true that the word εὐχαριστία is here used in the Didaché, and it is ordained that no unbaptized person shall eat of it. Yet I am disposed to believe the explanation to be, that the word Eucharist had not yet come to be used exclusively of the Lord's Supper. In the Clementines great prominence is given to Peter's benediction of meals in cases, where if an administration of the Eucharist, as we understand the word, be intended, Peter must have made every meal a Eucharist. For example, Clement, narrating his intercourse with Peter, previous to his baptism, says: And when he had said these things, and had taken food, he by himself, he commanded that I also should take food, and he blessed over the food, and gave thanks after he was satisfied,³⁷ and exhorted me with a word concerning that [which he had done]; and after these things he said, God grant thee that thou mayest in everything be like unto me, and mayest be baptized, and this same food with me thou mayest receive.³⁸

I do not know whether the influence of a Jewish original can be traced beyond chap. x.; and yet it is quite possible that a Jewish manual might contain directions as to the reception of ἀπόστσλοι, there being Jewish officers so called, as has been already remarked. And if the manual had contained orders as to the payment of first-fruits for the support of the high-priests, we could understand why the Didaché, in directing that first-fruits should be paid to the prophets, should add, for they are your high-priests. At any rate, chaps. xiv., xv., and the last chapter, on our Lord's Second Coming, are not likely to have had anything corresponding in a merely Jewish book. But there is one passage about which a few words must be said. I have said that in the section of Barnabas on the Two Ways there is no use of the Gospels; but there is one passage which apparently exhibits a use of the Acts and of St. Paul. Barnabas says (ch. xix.): Participate with your neighbour in all things, and say not that things are your own; for if you have been participators in that which is incorruptible, how much more in corruptible things. The passage strongly recalls Rom. xv. 27, and 1 Cor. ix. 11. But the same words are found both in the Didaché and in the Church Ordinances, save that instead of ἀφθάρτῳ we have ἀθανάτῳ. If we could take the three as independent witnesses, it would follow that there must have been corresponding words in the Jewish original; and then the question would arise whether that original may not have been old enough to have been known to St. Paul. But as there is also what looks like a use of Acts iv. 32, the passage can scarcely be pre-Christian; and I am therefore disposed to believe that Barnabas is here the original. I have already come to the conclusion that the Christian adapter of the Didaché" had seen Barnabas, and he may have made an addition from that source. I have not made any systematic study of the Church Ordinances; but I share the general belief that the latter half is not of the same date as the earlier portion;³⁹ and the later compiler may have been acquainted with the Didaché Some fuller account of the early use of the Didaché, and of the literature to which its discovery by Bryennius has given rise, will be found in an article (Teaching of the Twelve Apostles ) which I contributed to Smith and Wace's Dictionary of Christian Biography.

1) Lipsius, in his article APOCALYPSES, in Smith's Diet. Chr. Biog., states as on the authority of Eusebius (H.E. vi. 14), that Clement reckoned this Apocalypse among the antilegomena. But it was Eusebius, not Clement who so reckoned it. What the passage referred to says is, that Clement in his Hypotyposeis gave short comments (διηγήσεις) on all the Canonical Scripture, not even omitting the disputed books, viz. Jude and the other Catholic Epistles and the Epistle of Barnabas, and what is called the Apocalypse of Peter. With respect to Jude, see p. 493. Clement repeatedly quotes the Epistle of Barnabas, and appears to have no doubt of its Apostolic origin; and there is no reason to suppose that he thought less favourably of the Apocalypse of Peter.

2) The reader will note the Θεοῦ φύσις (see p. 555).

3) Many critics think that Macarius has preserved portions of a lost heathen work directed against Christianity: I now incline to the opinion that Macarius has exercised his rhetorical skill in writing the objections as well as the answers, though no doubt the objections were such as he had really encountered in controversy.

4) As it does not fall within my plan to treat of Old Testament Apocrypha, I content myself with mentioning that these Psalms are eighteen in number, and were probably written about 50 years before Christ. The list of the contents of Codex A shows that they formed part of that MS., following the Epistles of Clement; but these pages are now lost. These Psalms were edited from another MS. by Fabricius in his Codex Pseudep. V. T., and more recently by Hilgenfeld in his Messias Judaorum. In addition to the proof which the presence of these Psalms in Codex A affords that they obtained some amount of circulation among Christians, it may be mentioned that they are included in the Stichometry of Nicephorus, and that they are made use of in the Gnostic work, Pistis Sophia. That work contains several Psalms, some of which are adaptations of Psalms of David; others, of these Psalms of Solomon.

5) Lightfoot says (Clement, p. 12), Clement of Alexandria cites the "Apostle Clement" as he cites the "Apostle Barnabas," one of whose interpretations he nevertheless criticises and condemns with a freedom which he would not have allowed himself in dealing with writings regarded by him as canonical. I do not think that the passage referred to (Strom, ii. 15) quite warrants the inference drawn from it; and the phrase criticises and condemns is certainly too strong. Clement is engaged in showing that all sins are not equal, and he quotes, apparently with approbation, an exposition by Barnabas of the three classes of sinners referred to in Ps. i. i. It is scarcely a condemnation of Barnabas that he goes on to mention alternative, or even preferable, ways of making out the three classes. It is more to the purpose that Clement (Paed. ii. 10) corrects the natural history of Barnabas, but without mention of him by name.

6) Many of the Fathers have thought this exposition worth copying, e. g. Clem. Alex., Strom, vi. II, p. 782; Ambrose, De Abraha, i. 15; Prudentius, Psychom. 57; and even in our own times it has found a defender. Keble (Tracts for the Times, 89) says: In whatever measure the fact is made out, that the received Greek version of the Scriptures was under a peculiar providence, in the same degree it is rendered not improbable, that even in such an apparently casual thing as the number of Abraham's servants there was an eye to the benefit and consolation which the Church should long after receive, on recognizing, as it were, her Saviour's cypher, in the account of the one holy family triumphantly wrestling against the powers of the world. The Valentinians, whether deriving their method from Barnabas, or discovering it independently, found their 18 Aeons in the first two letters of the Saviour's name (Irenaeus I. iii. 2).

7) Westcott, for example, holds (N. T. Canon, p. 42) that Barnabas can in no case be ranked with the Twelve, or St. Paul, not having received his Apostolate directly from our Lord, as they did.

8) It is worth while, in this point of view, to compare this Epistle with the Gospel according to St. John, which has been characterized by some critics as anti- Jewish (see pp. 23, 277), but which will be seen to be in tensely Jewish as compared with Barnabas.

9) With regard to the suggestion, thrown out p. 457, that this may be the Epistle to the Alexandrians rejected, on account of its Marcionite tendencies, in the Muratorian Fragment, it must be borne in mind that even if our Epistle was really addressed to the Alexandrians, there is no evidence that it ever bore that title; and that it is even doubtful whether it was known in the West at the date of that Fragment.

10) A free quotation from Isaiah xlix. 17 (LXX.): ταχὺ οἰκοδομηθήσῃ ὑφ’ ὧν καθῃρέθης, καὶ οἱ ἐμημῶσαντές σε ἐξελεύσονται ἐκ σοῦ.

11) Not in the name of Clement, which is not once mentioned, and which we only learn to connect with the Epistle by independent tradition. In fact, it is remarkable how all through the first two centuries the importance of the bishop of Rome is merged in the importance of his Church. In the subsequent correspondence mentioned above, Dionysius of Corinth writes to the Church of Rome, not to Soter, its bishop. Ignatius, when on his way to suffer at the wild beast shows at Rome, writes to deprecate inter cession likely to be there made for his release; and he addresses the Church, not the bishop. And it is curious, that from this writer, who is accounted the strongest witness for Episcopacy in early times, we could not discover that there was any bishop at Rome. No mention is made of the bishop of Rome in the Shepherd of Hernias. And in the account which Epiphanius, evidently drawing from an older writer, gives of the intercourse of Marcion with the Church of Rome (Haer. 42), the dealings of Marcion are represented as being entirely with the Roman presbyters; and it may be doubted whether Epiphanius found in his authority the solution which he suggests, that at the time the see was vacant. At the very end of the century, when Victor attempted to enforce uniformity of Easter observance, it was still in the name of his Church that he wrote, asking that provincial councils should be assembled in order to report on the matter. This is evidenced by the plural ἠξιώσατε in the reply of Poly- crates (Euseb. v. 24).

12) This date has the authority of Eusebius (iii. 16), and apparently also the earlier authority of Hegesippus. What Eusebius says is, that in" the twelfth year of Domitian Clement succeeded to the bishopric of Rome; that he was the author of an admirable Epistle still extant, written in the name of the Church of Rome to the Church of Corinth, to appease a sedition in the latter Church; and that Hegesippus testifies that the sedition took place in the time of the afore -mentioned.

13) I make a suggestion in the next section as to the possible origin of the sedition.

14) The phrase is taken from the Prayer of Manasses, and seems to afford the earliest instance of its use. This document, which is included among the Apocryphal books of the Authorized Version, was not admitted into the Canon by the Council of Trent. But there is some evidence of early Church use of it. It is found in the Alexandrian MS., in the collection of hymns appended to the Psalter. It had been used by Julius Africanus (fr. 40, Routh, Rell. Sac. ii. 288), and it was copied into the Apostolic Constitutions, ii. 21.

15) ἐν τῷ ὑπό τινων καταφρονουμένῳ βιβλίῳ τῷ· Ποιμένι (De Princ. iv. 11).

16) Si non ab omni concilio ecclesiarum etiam vestrarum inter apocrypha et falsa judicaretur. We can infer from the vestrarum that the councils which condemned the Shepherd were later than the time of separation of Tertullian from the Church.

17) Ἰστέον ὣς καὶ τοῦτο πρὸς μὲν τινῶν ἆντιλέλεκται, δι’ οὓς οὒκ ἃν ἐν ἆμσλαγονμένως τεθείη, ὑφ’ ἑτέρων δὲ ἆνα-γκαιότατον οἷς μάλιστα δεῖ στυιχειῶσεως εἰσαγωγικῆς κέκριται, ὅθεν ἤδη καὶ ἐν ἐκκλησίαις ἴσμεν αἰπὺ δεδημοσιευμένον, καὶ τῶν παλαιοτάτων δὲ συγγραφέων κεχρημένους τινὰς αὐτῷ κατείληφα.

18) Having enumerated the books of Scripture, and declared these to be the only fountains of salvation, to which none may add nor take away, he goes "on to add, for greater accuracy, ὅτι ἔστι καὶ ἔτερα βιβλῗσ, τούτων ἔξωθεν, οὔ κανονιζόμενοι μὲν, τετυπωμένα. δὲ παρὰ τᾶν Πατέρων ἀναγινώσκεσθαι τοῖς ἄρτι προσερχομένου καὶ βουλομένοις κατηχεῖσθαι τὸν τῆς εὐσεβείας λόγον’ Σοφία Σολομῶντος καὶ Σοφία Σιρὰχ, καὶἘνθὴρ, καὶ Ἰονδὶθ, καὶ Τωβίας, καὶ Διδαχὴ καλουμένη τῶν Ἀτοιντόλων, καὶ ὁ Ποιμήν. And he proceeds to distinguish the two classes of books which he has enumerated from apocryphal books, which are only the invention of heretics.

19) The early date of Hermas was in recent times first seriously maintained by Zahn (Der Hirt des Hermas, 1868). Zahn is an authority whom it may not be safe always implicitly to follow, but who, at least, cannot be treated with disrespect. When he came forward to maintain the genuineness of the Ignatian letters he was regarded by many as the advocate of a hopeless cause; but Bishop Lightfoot's great work attests that he has won the verdict. I think he would have been more successful in gaining adherents in the present case, if the author with whom he deals were more generally read; for it appears to me that many scholars simply hold fast to the traditional opinion about a not very interesting book which they do not care to study for themselves. My own opinion was formed as the result of investigations commenced with a strong prepossession against the conclusion which I ultimately adopted.

20) On the method of solving historical difficulties by imagining for real characters duplicates unknown to history, the reader may consult's R. Maitland's tenth letter on Fox. If he does not know it already, he will thank me for the reference.

21) Smith's Diet. Chr. Biography, articles, MURATORIAN FRAGMENT, MONTANISM.

22) It was with much timidity that, in 1874, I named Hippolytus as probably the author of the work of which the Muratorian Fragment is a part; and hitherto my theory has found little acceptance. But I may now count Bishop Lightfoot as disposed to give his adherence. In a letter to the Academy (Sept. 21, 1889), he makes the luminous suggestion that the original of this Fragment was in verse, a supposition which throws great light on some characteristics of the document; and he justifies the suggestion by actual translation into Greek iambics of several passages of the Fragment. Now, among the works of Hippolytus, the titles of which are inscribed on his chair, is ᾠδαὶ εἰς πάσας τὰς γραφάς; and Lightfoot's idea is that we are to understand by this, verses on the Canon of O. T. and N. T. Scripture, of which verses the Fragment represents a part.

23) In Hermas, as in St. James's Epistle, the Christian community is ἡ ἐκκλησία, the assembly for worship, ἡ συναγωγή.

24) Those who take Hermas for a fictitious character are blind to the amusing little touches of human nature which constantly show themselves.

25) The claim is made in the subscription: Δανιὴλ κατὰ τοὺς ό. ἐγράφη ἐξ ἀντιγράφου ἔχοέτος τὴν ὑποσημείωιην ταύτην ἐγράφη ἐκ τῶν τετοιιπλῶν, ἐξ ὧνκαὶπαρετέθη.

26) This translation has been published by De Lagarde, Vet. Test, ab Origene recensiti fragmenta apud Syros servataquinque, Gottingen, 1880.

27) Accordingly, the Paschal Chronicle, following Epiphanius, places the publication of Theodotion's version in the consulship of Marcellus and Ælianus, that is, in the year 184.

28) Neander's main ground for suspicion (Antignosticus; ii. 530, Bohn) is that the treatise against the Jews has several passages in common with the third book against Marcion, which cohere with the context in the latter work, not in the former. It is clear, therefore, that the author of the former treatise borrowed these passages; but I hesitate to say that we can thence infer he was not Tertullian; for it is common with voluminous writers to save themselves trouble by turning to new account what they had written on a former occasion. I have myself pointed out (Hermathena, i. 103) that the use made (chap. 8) of the chronology of Hippolytus proves that the treatise against the Jews cannot be much earlier than A.D. 230, a time however when, there is reason to believe, Tertullian was still in literary activity.

Noeldechen, in his chronological arrangement of the writings of Tertul lian (Harnack, Texte und Untersuchungen, v. 2) places the treatise against the Jews quite early among the works of Tertullian (viz. about A.D. 195); but his reasons seem to me quite outweighed by those here given.

29) Dr. Gwynn has noted a verse (x. 6) in the LXX. Daniel, which affords ground for a suspicion that it was based on a former version, in points at least approaching to Theodotion's. There is nothing in the Hebrew corresponding to τὸ στόμα αὐτοῦ ὡσεὶ θαλάσσης, but this rendering might be accounted for as an editorial re-writing of τὸ σῶμα αὐτοῦ ὡσεὶ θάρσις, a literal rendering of the Hebrew preserved by Theodotion. The rendering of Tharshish by θάλασσα, though quite exceptional in the LXX., is found once Is. ii. 16, and has rabbinical authority: see also Jerome's Commentary in loc.; but it seems impossible to account for στόμα, except as a corruption of σῶμα. Dr. Gwynn observes also that the mistranslation σπανίσαι for seal (Dan. ix. 24) can scarcely be accounted for except as a corruption of the σφραγίσαι preserved by Theodotion.

30) Harnack calculates that the Didaché published by Bryennius would make 300 στίχοι.

31) See references in Lightfoot (Galatians, p. 92).

32) The Didaché fails to give any confirmation to the theory put forward by Dr. Hatch in his Bampton Lectures, that bishops and deacons were primarily appointed for the administration of the Church funds. Knowing that such administration was one of the bishop's functions in the time of Justin Martyr, we are rather surprised to find no mention in the Didaché that gifts intended for the poor passed through the hands of the bishops or deacons. Whatever may be meant by the gifts in Clem. Rom., ch. 44, the function there ascribed to the presbyters is that of offering, not of administering them; and the displaced Corinthian presbyters are com mended, not for the integrity with which they had discharged the latter office, but for the meekness with which they had borne their faculties in the former.

33) The Teaching of the Twelve Apostles; with Illustrations from the Talmud, by C. Taylor, D. D., Master of St. John's College, Cambridge. See also the Expositor, 3rd Series (1866), III. 316, 401. I am unable to notice other more recent Papers by Dr. Taylor and others, not having given any study to the Didaché since the publication of the last edition of this book.

34) This introduces a new element for the determination of the question (p. 569), whether or not the so-called Barnabas was a Jew. I now suspect that he had been a Gentile proselyte to Judaism, and had thus become acquainted with the Two Ways.

35) There is one Western quotation from Doctrines Apostolorum (pseudo- Cyprian, De Aleatoribus, p. 96, Hartel). It has affinities with a passage in Bryennius's Didaché, but differs a good deal in form. Harnack has lately ascribed the authorship of this tract De Aleatoribus to Victor of Rome; but for reasons which I cannot state here at length, I count it impossible that it can be quite so early.

36) There is, I think, reasonable ground to infer knowledge of the Didaché from one of the mysterious fragments, as from Irenaeus, published by Pfaff from a Turin Catena, which has since disappeared. I see no reason to doubt, that Pfaff found the extracts ascribed to Irenaeus in the MS. which he copied; but Catenae often make mistakes in their ascription of authorship, and though I believe the extract in question to have been from the work of an ancient author, I do not believe that that author was Irenaeus. Zahn's remark is conclusive, that this fragment quotes the Epistle to the Hebrews as St. Paul's.

37) Compare μετὰ τὸ ἐμπλησθῆναι (Didaché, ch. x.).

38) Clem. Recog. i. 19, translated for me from the Syriac by Dr. Gwynn. The strongest evidence that Clement of Alexandria knew the Palestinian form of the Didaché is, that he uses (Quts dives salvus, 20) the phrase vine of David, which occurs in one of these benedictory prayers. The phrase itself we may well believe occurred in the Jewish benediction, and there meant the Jewish people. And it is possible that this benediction may have been copied into the Egyptian form of the Apostolic Teaching. It is generally owned that the latter part of the Church Ordinances, as we have them, is a later addition; but in order to make room for that addition, the Way of Death, and possibly some other portions of the original document, have been cut away. Bornemann notices (Theol. Literaturz. 1885, 413) that Origen also has verae vitis quae ascendit de radice David (In Lib ruin Judicum Horn. 6, xi. 258, Lommatzsch).

39) There is in the latter one very curious passage (§ 26), indicating jealousy of the women on the part of the Apostles, which I suspect owes its origin to something in the Gospel according to the Egyptians. At least, the same feature shows itself in the Gnostic work, Pistis Sophia, which is also Egyptian. In p. 57, when Mary, who has already been highly commended by the Saviour for her previous answer, is about to speak, Peter leaps forward, and says: Lord, we cannot suffer this woman to take place with us, for she will not allow any of us to speak, but is speaking very often; and again, p. 161, Mary says: I would answer, but I am afraid of Peter, who is threatening me, and who hates our sex.