The following chapter was written by Samuel P. Tregelles, an eminent scholar of the text of the New Testament, who wrote this chapter before Tischendorf discovered the famous codex Sinaiticus, before the librarians of the Vatican had made codex Vaticanus fully accessible to scholars, before Westcott and Hort began their studies, and before the discovery of any of the papyrus manuscripts which figure so prominently in recent study. Yet it is remarkable to observe how Tregelles usually arrives at the same conclusions as later critical editors. This goes to show that in general the conclusions of recent editors do not depend upon a small number of recently-discovered manuscripts, nor upon any theory of recensions as developed by Westcott and Hort. Indeed, as Tregelles shows, these conclusions were anciently held by fathers of the church.

  Regarding Tregelles' treatment of Acts 20:28, it should be noted that his conclusion differs from that of some recent scholars. Codex Sinaiticus was found to support the reading, "church of God." But on the other hand, Tregelles' argument is now supported by Papyrus 74, which reads "church of the Lord"; and so both of these readings are usually indicated in modern versions.

  For the convenience of readers I have inserted within Tregelles' chapter some headings to divide it into sections. —M.D.M.

Notes on Some Passages of Dogmatic Importance

Samuel P. Tregelles, An Account of the Printed Text of the Greek New Testament; with Remarks on its Revision upon Critical Principles (London, 1854), pages 226-236.

Amongst the passages to the reading of which discussion has been directed on theological grounds, the more prominent are 1 John 5:7; 1 Tim 3:16; and Acts 20:28.

1 John 5:7

To enter into a formal discussion of the genuineness of the "testimony of the heavenly witnesses," 1 John 5:7, is really superfluous; for it would only be doing over again what has been done so repeatedly that there cannot be two opinions in the minds of those who now know the evidence, and are capable of appreciating its force. The passage stands thus (the words not known by the ancient authorities being enclosed within brackets): Verse 7, ὅτι τρεῖς εἰσιν οἱ μαρτυροῦντες [ἐν τῷ οὐρανῷ, ὁ Πατήρ, ὁ Λόγος, καὶ τὸ Ἅγιον Πνεῦμα· καὶ οὗτοι οἱ τρεῖς ἔν εἰσι. 8 καὶ τρεῖς εἰσιν οἱ μαρτυροῦντες ἐν τῇ γῇ] τὸ πνεῦμα καὶ τὸ ὕδωρ καὶ τὸ αἷμα, καὶ οἱ τρεῖς εἰς τὸ ἕν εἰσιν. I only add, that if the marked words be considered genuine, then any addition of any kind, found in any MS. (however recent), and supported by the later copies of any one version in opposition to the more ancient, possesses as good a claim to be received and used as a portion of Holy Scripture.

1 Timothy 3:16

In 1 Tim 3:16, there are three readings, θεος εφανερωθη εν σαρκι, as in the common text; ος εφαν. κτλ. and ο εφαν. κτλ. Now, to state the evidence for these readings respectively, it is necessary (as I had occasion long ago to point out), to divide the authorities at first into those which support the substantive θεος, and those which have in its stead a relative pronoun: what relative is the better supported by evidence is for after consideration.

In favour of the substantive. θεος is supported by the uncial MSS. J K (also D from a third corrector), and the cursive MSS. in general. 1 But it is upheld by no version whatever, prior to the Arabic of the Polyglot and the Sclavonic, both of which are more recent than the seventh century, and possess no value as critical witnesses. Some of the Greek fathers, who, as edited, have been cited as authorities for the reading θεος, ought to be omitted from the list; because it is certain, from other parts of their writings, that they did read ος in this passage, or because more exact collations of the MSS. of their works show that θεος is an unauthorized addition; so that in this case copyists have amplified by introducing this reading; just as in the former case they substituted it, as being that to which they were accustomed, for ος, which was then become peculiar.2

The fathers, then, who support θεος are, Didymus, Dionysius of Alexandria, and Theodoret, the two former possibly, the later not improbably; and in more recent times John Damascenus, Theophylact, and Ecumenius. Cyril Alex. and Chrysostom do not belong to this list.

In favour of a relative. ος is the reading of A C* F G, 17, and two other cursive MSS. ο is the reading of D*. It has, indeed, been said, that the true reading of A C F G is doubtful; and indeed, some have cited them all for θεος ; and it has been asserted also that G originally read ο.

Both A and C have suffered correction in this word; A in modern times, and C at a remote period. Such a change was effected by altering OC into ΘC [with a line drawn above] by introducing two little strokes, and then there was the contraction commonly found for θεος. The ink in which this has been done in A is sufficiently modern and black to declare its recent application, but it has been said that the trace of an original transverse line may be seen besides the modern black dot in the middle, decisive that the first letter is not O but Θ. Wetstein attributed this stroke, which in some lights is visible at one side of the O, to a part of the transverse line of the letter Ε on the back of the leaf. He says that it was only visible when he held it in such a position that he could see some light through the leaf. This was denied by Woide, who said (trusting to the eyes of others rather than his own) that the Ε was so placed that no part of it could be seen directly opposite to the O. Now I can state positively that Wetstein was right and Woide was wrong: for I have repeatedly looked at the place, sometimes alone, sometimes with others; sometimes with the unassisted eye, sometimes with the aid of a powerful lens: and as to the position of these two letters, by holding the leaf up to the light, it is seen that the Ε does slightly intersect the O, so that part of the transverse line may be seen on one side of that letter.

As to the reading of the palimpsest C, before the writing had been chemically restored, it was shown by Griesbach and others that the line denoting the contraction was not like the writing of the original copyist; and since the ancient letters have been revivified, it is abundantly manifest that both this stroke and the transverse line (previously invisible) forming the Θ are additions of a later corrector: Tischendorf states this explicitly in the Prolegomena to his edition of the text of this MS.; and I can abundantly confirm, from my own repeated inspection of the passage, and from comparing these strokes with the other corrections, that this is the fact.

With regard to F and G it is a mistake,3 that either or both of them read ΘC ; they read ος, and G has no correction in the place, as if it had ever read ο. It must be remembered that F and G are both of them copies of some one more ancient MS., and thus they are but one witness.

The versions which support a relative, are 1 the Old Latin, 2 the Vulgate, 3 Peshito and 4 Harclean Syriac, 5 Memphitic, 6 Thebaic, 7 Gothic, 8 Armenian, 9 Ethiopic: that is, all the versions older than the seventh century. (Also a MS. Arabic version in the Vatican.) This united testimony that θεος did not belong to the passages in the days when those versions were made, is peculiarly strong; and when it is remembered that no version of similar antiquity can be brought forward to counterbalance these witnesses of every region of Christendom, the preponderance of testimony is overwhelming.

It may now be stated that some of these versions cannot show whether they support ος or ο, from the want of genders in the relative; while others (such as the Vulgate), which mark the neuter, have given, not improbably, what was considered to be a constructio ad sensum, by taking μυστηριον as a personal designation for the antecedent. The two Syriac versions (the Harclean as to the text at least), the Armenian and the Ethiopic, are wholly doubtful as to this point: the Old Latin and the Vulgate have the neuter quod: the Gothic has the masculine relative, and so to the Memphitic and Thebaic; but, in the case of these two latter versions, it is said that the word by which μυστηριον is translated is also masculine, and so the masculine relative in itself proves nothing.

Theodorus of Mopsuestia, Cyril Alex., Epiphanius, read ος, while the Latin fathers in general (e.g. Hillary, Augustine, etc.) have quod. The silence of the fathers as to this passage in the fourth century, when, if they had known the reading θεος, it would have maintained an important part in arguments, must not be forgotten, for such silence expresses much.

In addition to the evidence of the MSS., versions, and early citations, there is a narrative which relates to this passage. According to this narrative, Macedonius, Patriarch of Constantinople, was deprived by the Emperor Anastasius, anno 506, for having corrupted the Scriptures (called in the account "evangelia," as a general term), especially in this passage, by changing one letter so as to make OC into ΘC.

"Hoc tempore Macedonius Constantinopolitanus episcopus ab imperatore Anastatio dicitur expulsus, tamquam evangelia falsasset, et maxime illud apostoli dictum, qui apparuit in carne, justficatus est in Spiritu. Hunc enim immutasse, obi habet ΟΣ, id est, QUI, monosyllabum Graecum; litera mutata O in Θ vertisse, et fecisse ΘΣ, id est, ut esset, DEUS apparuit per carnem. Tamquam Nestorianus ergo culpatus expellitur per Servum Monachum."

Such is the testimony of Liberatus Diaconus, 4 rather less than fifty years after the event took place. It has, indeed, been thought that the reading θεος could not have been introduced by one who was imbued with Nestorianism; for it has been said that this reading would contradict the distinction which that form of doctrine made between the natures of Christ, as though they were not joined in unity of person. But it must be remembered that Cyril was the orthodox authority then with the strong anti-Nestorian party, and he read ος εφανερωθη: also the reading θεος decidedly favoured the conception then formed of the doctrine of Nestorius; as if it had taught that God was manifest in or by the flesh of him who was born of Mary, whereas the reading ος strongly asserts unity of person.

This narration shows that in the early part of the sixth century the readings ος and θεος were both known; even if it be doubted whether this was the origin (as it may have been) of the latter. If it did so spring up,5 and if it was thus propagated, the versions made previously are witnesses against the addition: "cum multarum gentium linguis scriptura ante translata doceat falsa esse quae addita sunt," says Jerome (ad Damasum) of similar cases.

It is thus seen that for reading a relative pronoun in this place, there are MSS. A C D F G, 17, and two others, nine ancient versions, and some fathers certainly.

For reading the substantive θεος, there are J K (two of the later uncial MSS.), and the cursive copies in general; no version prior to the seventh century; and of the fathers of the earlier centuries there are only some doubtfully.

Codex B does not contain this epistle.

Thus the evidence in favour of a relative preponderates greatly: for it is not to be supposed that the independent more ancient versions could agree fortuitously in ignoring the substantive God, if they had it in their copies; and if none of them had it, then the Greek copies must have agreed in reading a relative.

The advocates for θεος, as being the reading supported by the numerical array of copies, are accustomed to divide the evidence into three heads, 1 θεος, 2 ος, 3 ο: and then, by giving the ancient versions in general to ο, they seem to make ος rest on weak grounds: but upon such a question the testimony of versions must not be separated thus minutely; for the primary question between the substantive and the relative must first be settled, just as in all preliminary inquiries, cognate readings must be taken as presenting united evidence, when contrasted with something wholly opposite.

A relative is then by far the best attested reading. The next inquiry is, what relative, ος or ο. This must be decided by Greek authorities, for most of the versions are doubtful. ος then has in its favour A C F G, 17, and two others, with Cyril and other Greek fathers, while ο is only supported by D a prima manu. Thus ος is by far the best supported reading.

It is also the reading from which the others might most easily have sprung from supposed correction; while the change from ο or θεος into ος would in such a sentence be most unlikely. And further, ος is the more difficult reading; for the inquiry immediately arises as to the structure and translation of the sentence: Does ος go back to θεου ζωντος for an antecedent? or are we to take μυστηριον ος for a constuctio ad sensum? or is the antecedent understood, that being the nominative to the verb of the next clause εδικαιωθη, "he who was manifested in the flesh, was justified," etc.? I do not think that either of these solutions is precisely the true one: ος appears to me to relate to the person indicated, with something of the same kind of indefinite emphasis (if I may use the term) as is found in the mode in which αυτος occurs in 1 John. "Confessedly great is the mystery of godliness: HE WHO was manifested in flesh, (he who) was justified in spirit, (he who) was seen by angels, (he who) was preached among Gentiles, (he who) was believed on in the world, (he who) was received up in glory."

The passage thus sets before us the whole dignity of Christ's person; and it has been well asked, if He were not essentially superhuman, how could the Apostle have emphatically declared that he was manifested in flesh?

Acts 20:28

I now pass on to Acts 20:28, ποιμαινειν την εκκλησιαν ... ην περιεποιησατο δια του αιματος του ιδιου. After εκκλησιαν there are three readings which are entitled to be considered as to their claims to fill up the place which I have left blank.
  1. Την εκκλησιαν του θεου, the church of God.
  2. Την εκκλησιαν του κυριου, the church of the Lord.
  3. Την εκκλησιαν του κυριου και θεου, the church of the Lord and God.
  There are also three readings which have to be mentioned simply with the evidence for them; none of which has a claim requiring much attention: (i) τ. εκκ. του κυριου θεου in one or two later MSS., and the Arabic of the Polyglot, a version of no critical importance; (ii) τ. εκκ. του θεου και κυριου, in one cursive copy; (iii) τ. εκκ. του χριστου as found in the Peshito Syriac (and of course in the Erpenian Arabic made from it); Origen so reads once; and this lection is found in three copies of Athanasius, and in Theodoret twice. It has no manuscript authority, and it might easily have sprung from the connection, in which the Church is mentioned as being his who redeeemed it with his own blood.

To revert, then, to the readings with regard to which there is some amount of evidence.

1. Του θεου. This is found in B, and about twenty cursive copies:6 and in the following versions (1) the Vulgate in the most ancient MSS., as well as in the common Clementine (but not, however, in the Complutensian edition). (2) the Harclean Syriac (text.), and a Syriac lectionary in the Vatican of the eleventh century. Epiphanius and some later Greek writers have this reading, as also have Ambrose and other Latins. Athanasius in some MSS. has this reading, and Chrysostom has been cited for it; however, he certainly himself has κυριου, and the reading θεου has been taken from the Homilies on the Acts which bear his name; but even there the reading is doubtful.7 Cyril of Alexandria reads θεου twice, in a treatise on the name θεοτοκος, as applied to the Virgin Mary, edited by Cardinal Mai (Scriptorum Collectio Vaticana, viij. part 2, pp. 125, 126). It is necessary to notice this explicitly, because it has been remarked that this reading is not found in Cyril, and the supposed silence of this anti-Nestorian writer has been made the basis of argument. The genuineness of this treatise is supported by its being cited in the Emperor Justinian's epistle to the Alexandrian monks (p. 306), edited by Mai in vol. vii. of the same collection. This treatise is likewise thoroughly Cyrillian in tone and style. 8

2. Του κυριου is the reading of A C D E, 13 (with thirteen other cursive MSS.), of (1) the Old Latin, as found in D and E, (2) the Memphitic, (3) the Thebaic, (4) the Armenian, and (5) the margin of the later Syriac. Irenaeus (or his contemporary Latin interpreter), Eusebius, the Apostolic Constitutions, Didymus, Ammonius, Athanasius in one MS., Chrysostom (on Eph. 4:12), and at a later date Theophylact (three times), have this reading; as also, among the Latins, Lucifer, Jerome, Augustine, and others.

3. Του κυριου και θεου: this is the common reading of MSS., being found in G H, (also C a tertia manu) and in more than a hundred cursive copies, also in six lectionaries. As to versions, it is found in the Sclavonic alone, 9 which is of the ninth century, and has no voice in criticism. Theophylact has this reading once, so that when he has του κυριου simply, he may probably abbreviate the reading to which he was accustomed. This reading is found in the Complutensian edition, and as it is that supported by numbers, it would of course have been defended by many if it had been in the common text. The Latin in the Complutensian differs from other copies of the Vulgate in having "dni (i.e. Domini) et Dei."

In this conspectus of authorities, the Ethiopic version has not been cited for any of the readings: it is doubtful whether the Roman text of this version should be quoted for θεου or κυριου, and the edition of Mr. Platt has χριστου. All that can be said is, that, like the Peshito Syrac, it opposes the compound reading του κυριου και θεου.

The whole question must lie between του κυριου and του θεου; for the reading that combines both fails as to ancient MS. authority (showing plainly that the mass of copies must not be valued on the ground of numbers), as to versions, and as to early citations: if this had not been sufficient, it might be added that it is the longer reading, and as such would require preponderating evidence before it could be received.

Του θεου has good witnesses in B (the other MSS. are unimportant) and the Vulgate; but του κυριου has preponderating testimony; for B alone could not on such a point outweigh A C D E; and as to versions and fathers, του κυριου stands on stronger ground; and therefore it should be accepted, even while all that can be said in favour of του θεου is fully admitted. Either of these readings might easily have sprung from the other, as the change is but one letter (ΚΥ and ΘΥ [with a stroke above to indicate abbreviation]); and while θεου might claim the preference as being, in connection with "blood," the more difficult reading, η εκκλησια του κυριου is a reading found nowhere else in the New Testament; so that a copyist would naturally alter it to εκκ. του θεου, as is found in 1 Cor. 1:2, 10:32, 11:22, 15:9; 2 Cor. 1:1; Gal. 1:13; 1 Tim. 3:5, 15. This whole passage may also be compared with 1 Pet 5:2, ποιμανατε το εν υμιν ποιμνιον του θεου επισκοπουντες, which might aid in suggesting του θεου in Acts 20:28, προσεχετε ... τω ποιμνιω εν ω υμας το πνα το αγιον εθετο επισκοπους ποιμαινειν την εκκλησιαν του κυριου. Thus the introduction of θεου instead of κυριου would be natural, though the contrary would not be so; and even if the evidence for εκκ. του κυριου had not been so strong, it would have been confirmed by its peculiarity, and by the immense probability of the familiar phrase being substituted for it.

But although this passage with the reading κυριου gives no direct testimony to the Godhead of the Lord Jesus Christ, it is of very great doctrinal value; for it brings out in full view the true sacrificial character of his death on the cross: "Feed the church of the Lord, which He hath purchased with his own blood." Thus, even if the dignity of his person were not here stated, the preciousness of his blood is emphatically declared, as being that which was adequate to meet the infinite holiness of God and his wrath against sin, and to secure the church unto Christ as his own, as that which he has appropriated at so costly a price. If this work of propitiation is rightly considered, and its value as thus declared as applied in result, how much does it show that the dignity of this Redeemer exceeds that of a mere man. His blood was so unspeakably precious that it was capable of outweighing, even before God, the sins of all his people; and this it is that shows how exalted must be the person of whom such things could be spoken. If this passage, as rightly read, does not declare our Lord's Godhead, it still states, in clearest words, his redemption and Lordship.

Sure Testimonies

Many have shrunk from the results of criticism because of these three passages: they are accustomed to them as setting forth theological verities; and they have desired to cling to them; although they might have known that in argument they are worthless, because opposers are full well aware how groundless or uncertain are those readings of these passages which some have called orthodox. The consequence unhappily has been, that the most essential and fundamental truths of Christian doctrine have been supposed by some to rest on uncertain grounds. Now, the same criticism which shows that particular readings are not genuine, proves uncontestably that others are unquestionable; and thus no point of orthodox truth is weakened, even though supports, which some have thought sustained it, are found to differ from such supposed use and bearing. There are undoubted passages enough (such as Matt. 1:23; John 1:1, 20:28; Rom. 9:5; Phil. 2:6; Heb. 1:8) which speak of the proper Godhead of Christ, without our wishing to press into the same cause others for which we have no sufficient evidence, and which were not required to establish that necessary truth in the early controversies.

Criticism, however, need not be at all feared; if it takes away on the one hand readings which were thought to have some dogmatic value, it will give on the other quite as much. Instances of this will be seen in two passages, John 1:18, and 1 Pet. 3:15.

John 1:18

John 1:18, θεον ουδεις εωρακεν πωποτε ο μονογενης υιος ο ων εις τον κολπον του πατρος εκεινος εxηγησατο.

Here, instead of μονογενης υιος of the common text, great authorities support μονογενης θεος. This is the reading of B C* L, 33. (As to B, this reading is given in Bartolocci's MS. collation at Paris, and I myself saw it in the MS. at Rome; in C it was chemically brought to light.) This is supported by the following versions, the Peshito Syriac, and the margin of the Harclean; the Memphitic (sic) and the Ethiopic: and as to fathers, the reading may almost be called general, for it is that of Clement of Alexandria, Irenaeus, Origen, Eusebius, Epiphanius, Lucian, Basil, Gregory of Nazianzum, Gregory of Nussa, Didymus, Basil of Seleucia, Isadore of Pelusium, Cyril of Alexandria, Titus of Bostra; as also of Theodotus (in the second century), Arius, Marcellus, Eunomius, etc. ; and amongst the Latins, Hilary, Fulgentius, Gaudentius, Ferrandus, Phoebadius, Vigilius, Alcuin, etc. The reading of the common text, υιος, is found in A and the MSS. in general: of these A alone belongs to the most ancient class; D is here defective. It is that of the Old Latin, of the Vulgate, the Curetonian Syriac, the text of the Harclean Syriac, and the Jerusalem Syriac Lectionary, and the Armenian. It is found twice in Origen, in Eusebius, Basil, and Irenaeus (though all these writers have also the other reading, and in general they so speak of θεος in the passage, that υιος must have proceeded from the copyists): - the Latin writers in general agree with the Latin versions in reading filius.

In forming a judgment between these two readings, it must be remembered that μονογενης would naturally suggest υιος as the word which should follow it, whereas θεος strikes the ear as something peculiar, and not elsewhere occuring in Scripture; the change, being but of one letter (ΥC for ΘC [with a stroke above to indicate abbreviation]), might be most inadvertently made; and though the evidence of the Latin versions and the Curetonian Syriac is not of small weight, yet the same chance of change would, in a case of this kind, affect the copyists of a version (or indeed the translators) just as much as the transcribers of Greek MSS. Θεος, as the more difficult reading, is entitled to especial attention; and, confirmed as it is by MSS. of the highest character, by good versions, and by the general consent of the early Greek writers (even when, like Arius, they were opposed to the dogma taught), it is necessary, on grounds of combined evidence, to receive it in preference to the easier and more natural reading υιος. No critical edition hitherto published has given θεος in the text; it is placed, however, in Lachmann's inner margin, as a reading between which and that in the text the evidence stands in doubt: he gave it that place on the combined testimony of Origen and Irenaeus, but he did not know (for then it was not ascertained) that this reading is that of B and C, two of the principal witnesses that he admitted. 10

1 Peter 3:15

1 Pet. 3:15, κυριον δε τον θεον αγιασατε, so the common text; but instead of θεον the reading χριστον is supported by most preponderating evidence; for it is the reading of A B C, 13, and some other cursive MSS.; of the Vulgate, the Peshito and Harclean Syriac, the Memphitic, Thebaic, Armenian (the Ethiopic has neither word); it is also cited by Clement and others: the reading θεον is supported by the evidence of no MS. older than G and J (at Moscow) of the ninth century, and it is found in no version older than the Arabic in the Polyglot. Thus the reading χριστον may be relied on confidently. This occurs in a citation by the Apostle from Isa. 8:12,13. In the Prophet the words are, "Neither fear ye their fear nor be afraid; sanctify the Lord of Hosts himself." The citation of the Apostle exactly agrees with this, except in the concluding words, in which, in the corrected text we have κυριον δε τον χριστον αγιασατε, "sanctify the Lord Christ" : this shows that the expression "Jehova of Hosts himself" in the Prophet, finds its New Testament exposition as an equivalent in κυριον τον χριστον, "the Lord Christ," thus marking the divine glory of our Lord in the most emphatic manner. And this is in thorough accordance with the Apostle's train of thought; for the following words of the Prophet, in which he says that Jehovah of Hosts should become "a stone of stumbling and rock of offence," had been previously applied by him (chap. 2:7,8) to the Lord Jesus. The LXX., which so often has influenced copyists to bring passages in the New Testament into verbal conformity with it, has not caused the introduction of the word θεον; for the passage there runs, τον δε φοβον αυτου ου μη φοβηθητε ουδε μη ταραχθητε κυριον αυτον αγιασατε. In this citation the Apostle shows how independent the New Testament writers can be of the LXX. when needful; indeed, in some part of the passage the LXX. so reads as utterly to contradict both the Hebrew text and the New Testament use of the facts previously revealed. To the LXX. translators it was incomprehensible that the Lord could become a stone of stumbling and rock of offence to Israel; and thus, in verse 14, a negative is introduced, και ουχ ως λιθον προσκομματι συναντησεσθε ουδε ως πετρας πτωματι. On such points, and all that relate to the Godhead of Christ, and in doctrinal statements, the LXX. is continually at variance with both the New Testament and the Hebrew text.


1. In one cursive MS., Cod. Leicest., I observed that the reading is ο θεος.

2. Thus Cyril Alex. really read ος, though in his printed works θεος also occurs; the very context would prove that this latter reading had no place in Cyril's sentence. Several MSS. contain a scholion to the purport that ος was the Cyrillian reading, even though the MSS. themselves contain the common text θεος (ο εν αγιοις κυριλλος ... φησιν, ΟΣ εφανερωθη).
   Chrysostom has been cited in favour of θεος ; but I have had occasion to point out that though the word so stands in the editions, yet the citation of the same passage of Chrysostom in the Catena on 1 Tim., published by Cramer (p. 31), shows plainly that εις ετερον αναγει το πραγμα. οτι εφανερωθη εν σαρκι, has been transformed into εις ετερον αναγει το πραγμα, λεγων θεος εφανερωθη εν σαρκι.

3. See above p. 165, note.

4. Breviarum, cap. xix. I take the citation from Bently (Dyce's edition, iij. 366), who adds, "The editions of Liberatus, instead of Θ and ΘΣ, have Ω and ΩΣ; but it appears from Baronius, that the manuscript had no Greek letters here at all, and that they were supplied by the first editor. I have not scrupled, therefore, to correct the place, as the Latin clearly requires: for DEUS answers to ΘΕΟΣ, and the Greek monosyllable ΟΣ is in opposition to that dissyllable. And so Hinemarus in his Opusculum, chap. xviij., where he cites the same story (without doubt out of Liberatus), has it plainly, as I have put it, O in Θ vertit et fecit ΘΣ." It is important to remember this fact out of Baronius, that the MS. of Liberatus had no Greek letters; for it has been cited again and again, as if it had been said that Macedonius changed ος into ως, and this has even been put in opposition to the testimony of Hinemar. "The first editor," whoever he may have been, had probably the same notion how a short O might be interchanged with a long one, and hence the mistake; - one which might have been avoided, if he had noticed the Latin qui and Deus; but probably he did not understand that ΘΣ would be the common contraction for θεος.
  The same transaction regarding Macedonius and the corruption of Scripture is referred to in the Chronicon of Victor. "Messala V.C. Cross. Constantinopoli, jubente Anastasio imperatore, sancta evangelia tamquam ab idiotis evangelistis composita, reprehenduntur et emendantur."

5. If so, the occurrence of θεος in any earlier citations must be occasioned by copyists or editors assimilating, pro more, the Biblical citations to the text which they were accustomed to read.

6. As doubt has been cast on the reading of B, I state explicitly that this is the reading of that MS. The late Mr. Edgar Taylor procured a tracing of rather more than three lines in this passage from the custode of the Vatican Library: and it appeared in the editorial Monitum prefixed to the second London reprint of Griesbach's Greek Testament (1818). But it was soon suggested that though the MS. now reads ΘΥ, it might formerly have had ΚΥ : I therefore, when at Rome, directed my attention particularly to that point, and I can state positively that the Θ stands without any erasure, or trace of there having been originally a Κ. This was contrary to what I had expected; for I had quite anticipated that I should have found that it had at first the same reading as A C.

7. In expressing my opinion that the Homilies on the Acts are not really Chrysostom's, I shall not be accused of rashness by those who understand the real state of the question; a statement which I once made that I thought they were not really his, was met by such remarks as if this was some new opinion of my own, previously maintained by no one. In reading those Homilies, I felt often astonished at their contents and style being so un-Chrysostomlike; and this was when I had for some weeks read hardly anything except his works, so that my perceptions were fully alive as to such points. On examination I found that, from Erasmus onwards, scholars had doubted or denied that this work is genuine. This was no small confirmation of my previously formed judgment.

8. Some of the other works published by Mai in the same place as Cyril's, are certainly not his (in one of these, p. 56, κυριου is cited in this passage); they contain abundant proof that they were subsequent to the Eutychian controversy; and not only do they combat heresy of later date than Cyril's time, but they express sentiments by no means Cyrillian.

9. It is instructive to see how repeatedly, when the mass of modern MSS. oppose the ancient, they are supported by no versions except those later than the seventh century. In speaking of the Sclavonic as belonging to the ninth century, I do not discuss whether or not the other books were translated about the same time as the Gospels. We know when this version was began, but as to its completion we have no evidence: the oldest existing MS. of the whole Sclavonic Bible is of A.D. 1499. (Davidson's Biblical Criticism, ii. p.238.)

10. When Lachmann really knew from me the MS. authority in favour of θεος, he at once admitted the claim of that word to stand in the text instead of υιος. Indeed, his principal witness for giving the preference to the latter word was B, which had been supposed to read thus.