Let us now draw up a table of Justin's quotations from the Gospels arranged as nearly as may be on the same standard and scale as that of the quotations from the Old Testament. Such a table will stand thus. [Those only which appear to be direct quotations are given.]

Exact. Slightly variant. Variant. Remarks. D.49, Matt. 3.11, repeated in part 12 (v.l.) similarly. D. 51, Matt. 11. compounded with 12-15; Luke 16. omissions but 16. striking resem- blances. D. 49, Matt. 17. 11-13. A.1.15, Matt. 5.28. A.1.15, Matt. 5. from memory? 29; Mark 9.47. A.1.15, Matt. 5.32. confusion of read- ings. A.1.15, Matt. from memory? 19.12. A.1.15, Matt. 5. compounded. 42; Luke 6.30, 34. Continuous.{ A.1.15, Matt. 6. { 19, 20; 16.26; 6.20. Continuous.{ A.1.15 (D.96), from memory(Cr.), { Luke 6.36; but prob. diff- { Matt. 5.45; 6. erent document; { 25-27; Luke 12. rather marked { 22-24; Matt. 6. identity in { 32, 33; 6.21. phrase. A.1.15, Matt. 6.1. A.1.15, Matt. 9. do the last 13(?). words belong to the C quotation? o { A.1.15, Luke n { 6.32; Matt. t { 5.46. i { A.1.15, (D. repeated in part n { 128), Luke similarly, in u { 6.27, 28; part diversely; o { Matt. 5.44. confusion in u MSS. s s Continuous. { A.1.16, Luke 6.29 { (Matt. 5.39, 40.) { A.1.16, Matt. 5. { 22 (v.l.) { A.1.16, Matt. 5 [Greek: { 41. angaeusei.] { A.1.16, Matt. 5.16. D.93, A,1.16, Matt. 22.40,37, 38. A.1.16, D.101, repeated Matt. 19.16, diversely. 17 (v.l.); Luke 18.18,19 (v.l.) A.1.16, Matt. 5. 34,37. {A.1.16, Matt. { 7.21. C { A.1.16 (A.1.62), repeated in part o { Luke 10.16 (v.l.) similarly, in n { part diversely. t { A.1.16 (D.76), i { Matt. 7.22, 23 n { (v.l.); Luke u { 13.26,27 (v.l.) o { A.1.16, Matt. 13. addition. u { 42, 43 (v.l.) s { A.1.16 (D.35), { Matt. 7.15. { A.1.16, Matt. 7. { 16, 19. D.76, Matt. 8.11. 12+. D.35, [Greek: esontai schi- smata kai hai- reseis.] D.76, Matt. 25.41 (v.l.) D.35, Matt. 7.15. repeated with nearer approach to Matthew, perh. v.l. D.35, 82, Matt. repeated with 24.24 (Mark 13. similarity and 22). divergence. D.82, Matt. 10. freely. 22, par. A.1.19, Luke 18. 27+. A.1.19, Luke 12. compounded. 4, 5; Matt. 10.28. A.1.17, Luke 12. 48 (v.l.) D.76, Luke 10.19+ ins. [Greek: skolopendron.] D.105, Matt. 5. 20. D.125, Matt. 13. condensed narra- 3 sqq. tive. +D.17, Luke 11. 52. D.17, Matt. 23.23; compounded. Luke 11.42. D.17, 112, Matt. repeated simi- 23.27; 23.24. larly. D.47, [Greek: en ois an humas katalabo en toutois kai krino.] D.81, Luke 20. marked resem- 35, 36. blance with difference. D.107, Matt.16.4. D. 122, Matt. 23. 15. D.17, Matt. 21. 13, 12. A.1.17, Luke 20. narrative portion 22-25 (v.l.) free. D.100, A.1.63, repeated not Matt. 11.27 (v.l.) identically. D.76, 100, Luke repeated diverse- 9.22. diversely; free (Credner). A.1.36, Matt. 21. D.53, Matt. 21.5. (Zech. 9.9). 5 (addition). A.1.66, Luke 22. 19, 20. D.99, Matt. 26. 39 (v.l.) D.103, Luke 22. 42-44. D.101, Matt. 27. 43. A.1.38, [Greek: ho nekrous anegeiras rhu- sastho eauton.] D.99, Matt. 27. compounded. 46; Mark 15.34. D.105, Luke 23. 46.

The total result may be taken to be that ten passages are substantially exact, while twenty-five present slight and thirty- two marked variations [Endnote 116:1]. This is only rough and approximate, because of the passages that are put down as exact two, or possibly three, can only be said to be so with a qualification; though, on the other hand, there are passages entered under the second class as 'slightly variant' which have a leaning towards the first, and passages entered under the third which have a perceptible leaning towards the second. We can therefore afford to disregard these doubtful cases and accept the classification very much as it stands. Comparing it then with the parallel classification that has been made of the quotations from the Old Testament, we find that in the latter sixty-four were ranked as exact, forty-four as slightly variant, and fifty-four as decidedly variant. If we reduce these roughly to a common standard of comparison the proportion of variation may be represented thus: Exact. Slightly Variant. variant. Quotations from the Old Testament 10 7 9 Quotations from the Synoptic Gospels 10 25 32

It will be seen from this at once how largely the proportion of variation rises; it is indeed more than three times as high for the quotations from the Gospels as for those from the Old Testament. The amount of combination too is decidedly in excess of that which is found in the Old Testament quotations.

There is, it is true, something to be said on the other side. Justin quotes the Old Testament rather as Scripture, the New Testament rather as history. I think it will be felt that he has permitted his own style a freer play in regard to the latter than the former. The New Testament record had not yet acquired the same degree of fixity as the Old. The 'many' compositions of which St. Luke speaks in his preface were still in circulation, and were only gradually dying out. One important step had been taken in the regular reading of the 'Memoirs of the Apostles' at the Christian assemblies. We have not indeed proof that these were confined to the Canonical Gospels. Probably as yet they were not. But it should be remembered that Irenaeus was now a boy, and that by the time he had reached manhood the Canon of the Gospels had received its definite form.

Taking all these points into consideration I think we shall find the various indications converge upon very much the same conclusion as that at which we have already arrived. The a priori probabilities of the case, as well as the actual phenomena of Justin's Gospel, alike tend to show that he did make use either mediately or immediately of our Gospels, but that he did not assign to them an exclusive authority, and that he probably made use along with them of other documents no longer extant.

The proof that Justin made use of each of our three Synoptics individually is perhaps more striking from the point of view of substance than of form, because his direct quotations are mostly taken from the discourses rather than from the narrative, and these discourses are usually found in more than a single Gospel, while in proportion as they bear the stamp of originality and authenticity it is difficult to assign them to any particular reporter. There is however some strong and remarkable evidence of this kind.

At least one case of parallelism seems to prove almost decisively the use of the first Gospel. It is necessary to give the quotation and the original with the parallel from St. Mark side by side.

Justin, Dial. c.49.

[Greek: Aelias men eleusetai kai apokatastaesei panta, lego de humin, hoti Aelias aedae aelthe kai ouk epegnosan auton all' epoiaesan auto hosa aethelaesan. Kai gegraptai hoti tote sunaekan oi mathaetai, hoti peri Ioannon tou Baptistou eipen autois.]

Matt. xvii. 11-13.

[Greek: Aelias men erchetai apokatastaesei panta, lego de humin hoti Aelias aedae aelthen kai ouk epegnosan auton, alla epoiaesan auto hosa aethelaesan, [outos kai ho uios tou anthropou mellei paschein hup' auton.] Tote sunaekan oi mathaetai hoti peri Ioannou tou Baptistou eipen autois.] The clause in brackets is placed at the end of ver. 13 by D. and the Old Latin.

Mark. ix. 12, 13.

[Greek: Ho de ephae autois, Aelias [men] elthon proton apokathistanei panta, kai pos gegraptai epi ton uion tou anthropou, hina polla pathae kai exoudenaethae. Alla lego humin hoti kai Aelias elaeluthen kai epoiaesan auto hosa aethelon, kathos gegraptai ep' auton.]

We notice here, first, an important point, that Justin reproduces at the end of his quotation what appears to be not so much a part of the object-matter of the narrative as a comment or reflection of the Evangelist ('Then the disciples understood that He spake unto them of John the Baptist'). This was thought by Credner, who as a rule is inclined to press the use of an apocryphal Gospel by Justin, to be sufficient proof that the quotation is taken from our present Matthew [Endnote 119:1]. On this point, however, there is an able and on the whole a sound argument in 'Supernatural Religion' [Endnote 119:2]. There are certainly cases in which a similar comment or reflection is found either in all three Synoptic Gospels or in two of them (e.g. Matt. vii. 28, 29 = Mark i. 22 = Luke iv. 32; Matt. xiii. 34 = Mark iv. 33, 34; Matt. xxvi. 43 = Mark xiv. 40; Matt. xix. 22 = Mark x. 22). The author consequently maintains that these were found in the original document from which all three, or two Synoptics at least, borrowed; and he notes that this very passage is assigned by Ewald to the 'oldest Gospel.'

The observation in itself is a fine and true one, and has an important bearing upon the question as to the way in which our Synoptic Gospels were composed. We may indeed remark in passing that the author seems to have overlooked the fact that, when once this principle of a common written basis or bases for the Synoptic Gospels is accepted, nine-tenths of his own argument is overthrown; for there are no divergences in the text of the patristic quotations from the Gospels that may not be amply paralleled by the differences which exist in the text of the several Gospels themselves, showing that the Evangelists took liberties with their ground documents to an extent that is really greater than that of any subsequent misquotation. But putting aside for the present this argumentum ad hominem which seems to follow from the admission here made, there is, I think, the strongest reason to conclude that in the present case the first Evangelist is not merely reproducing his ground document. There is one element in the question which the author has omitted to notice; that is, the parallel passage in St. Mark. This differs so widely from the text of St. Matthew as to show that that text cannot accurately represent the original; it also wants the reflective comment altogether. Accordingly, if the author will turn to p. 275 of Ewald's book [Endnote 120:1] he will find that that writer, though roughly assigning the passage as it appears in both Synoptics to the 'oldest Gospel,' yet in reconstructing the text of this Gospel does so, not by taking that of either of the Synoptics pure and simple, but by mixing the two. All the other critics who have dealt with this point, so far as I am aware, have done the same. Holtzmann [Endnote 120:2] follows Ewald, and Weiss [Endnote 120:3] accepts Mark's as more nearly the original text.

The very extent of the divergence in St. Mark throws out into striking relief the close agreement of Justin's quotation with St. Matthew. Here we have three verses word for word the same, even to the finest shades of expression. To the single exception [Greek: eleusetai] for [Greek: erchetai] I cannot, as Credner does [Endnote 120:4], attach any importance. The present tense in the Gospel has undoubtedly a future signification [Endnote 120:5], and Justin was very naturally led to give it also a future form by [Greek: apokatastaesei] which follows. For the rest, the order, particles, tenses are so absolutely identical, where the text of St. Mark shows how inevitably they must have differed in another Gospel or even in the original, that I can see no alternative but to refer the quotation directly to our present St. Matthew.

If this passage had stood alone, taken in connection with the coincidence of matter between Justin and the first Gospel, great weight must have attached to it. But it does not by any means stand alone. There is an exact verbal agreement in the verses Matt. v. 20 ('Except your righteousness' &c.) and Matt. vii. 21 ('Not every one that saith unto me,' &c.) which are peculiar to the first Gospel. There is a close agreement, if not always with the best, yet with some very old, text of St. Matthew in v. 22 (note especially the striking phrase and construction [Greek: enochos eis]), v. 28 (note [Greek: blep. pros to epithum].), v. 41 (note the remarkable word [Greek: angareusei]), xxv. 41, and not too great a divergence in v. 16, vi. 1 ([Greek: pros to theathaenai, ei de mae ge misthon ouk echete]), and xix. 12, all of which passages are without parallel in any extant Gospel. There are also marked resemblances to the Matthaean text in synoptic passages such as Matt. iii. 11, 12 ([Greek: eis metanoian, ta hupodaemata bastasai]), Matt. vi. 19, 20 ([Greek: hopou saes kai brosis aphanizei], where Luke has simply [Greek: saes diaphtheirei], and [Greek: diorussousi] where Luke has [Greek: engizei]), Matt. vii. 22, 23 ([Greek: ekeinae tae haemera Kurie, Kurie, k.t.l.]), Matt. xvi. 26 ([Greek: dosei] Matt. only, [Greek: antallagma] Matt., Mark), Matt. xvi. 1, 4 (the last verse exactly). As these passages are all from the discourses I do not wish to say that they may not be taken from other Gospels than the canonical, but we have absolutely no evidence that they were so taken, and every additional instance increases the probability that they were taken directly from St. Matthew, which by this time, I think, has reached a very high degree of presumption.

I have reserved for a separate discussion a single instance which I shall venture to add to those already quoted, although I am aware that it is alleged on the opposite side. Justin has the saying 'Let your yea be yea and your nay nay, for whatsoever is more than these cometh of the Evil One' ([Greek: Mae omosaete holos. Esto de humon to nai nai, kai to ou ou; to de perisson touton ek tou ponaerou]), which is set against the first Evangelist's 'Let your conversation be Yea yea, Nay nay, for whatsoever is more than these cometh of the Evil One' ([Greek: ego de lego humin mae omosai holos... Esto de ho logos humon nai nai, ou ou; to de perisson, k.t.l.]). Now it is perfectly true that as early as the Canonical Epistle of James (v. 12) we find the reading [Greek: aeto de humon to nai nai, kai to ou ou], and that in the Clementine Homilies twice over we read [Greek: esto humon to nai nai, (kai) to ou ou], [Greek: kai] being inserted in one instance and not in the other. Justin's reading is found also exactly in Clement of Alexandria, and a similar reading (though with the [Greek: aeto] of James) in Epiphanius. These last two examples show that the misquotation was an easy one to fall into, because there can be little doubt that Clement and Epiphanius supposed themselves to be quoting the canonical text. There remains however the fact that the Justinian form is supported by the pseudo-Clementines; and at the first blush it might seem that 'Let your yea be yea' (stand to your word) made better, at least a complete and more obvious, sense than 'Let your conversation be' (let it not go beyond) 'Yea yea' &c [Endnote 122:1]. There is, however, what seems to be a decisive proof that the original form both of Justin's and the Clementine quotation is that which is given in the first Gospel. Both Justin and the writer who passes under the name of Clement add the clause 'Whatsoever is more than these cometh of evil' (or 'of the Evil One'). But this, while it tallies perfectly with the canonical reading, evidently excludes any other. It is consequent and good sense to say, 'Do not go beyond a plain yes or no, because whatever is in excess of this must have an evil motive,' but the connection is entirely lost when we substitute 'Keep your word, for whatever is more than this has an evil motive'—more than what?

The most important points that can be taken to imply a use of St. Mark's Gospel have been already discussed as falling under the head of matter rather than of form.

The coincidences with Luke are striking but complicated. In his earlier work, the 'Beitraege' [Endnote 123:1], Credner regarded as a decided reference to the Prologue of this Gospel the statement of Justin that his Memoirs were composed [Greek: hupo ton apostolon autou kai ton ekeinois parakolouthaesanton]: but, in the posthumous History of the Canon [Endnote 123:2], he retracts this view, having come to recognise a greater frequency in the use of the word [Greek: parakolouthein] in this sense. It will also of course be noticed that Justin has [Greek: par. tois ap.] and not [Greek: par. tois pragmasin], as Luke. It is doubtless true that the use of the word can be paralleled to such an extent as to make it not a matter of certainty that the Gospel is being quoted: still I think there will be a certain probability that it has been suggested by a reminiscence of this passage, and, strangely enough, there is a parallel for the substitution of the historians for the subject-matter of their history in Epiphanius, who reads [Greek: par. tois autoptais kai hupaeretais tou logou] [Endnote 124:1], where he is explicitly and unquestionably quoting St. Luke.

There are some marked coincidences of phrase in the account of the Annunciation—[Greek: eperchesthai, episkaizein, dunamis hupsistou] (a specially Lucan phrase), [Greek: to gennomenon] (also a form characteristic of St. Luke), [Greek idou, sullaepsae en gastri kai texae huion]. Of the other peculiarities of St. Luke Justin has in exact accordance the last words upon the cross ([Greek: Pater, eis cheiras sou paratithemai to pneuma mou]). In the Agony in the Garden Justin has the feature of the Bloody Sweat; but it is right to notice—

(1) That he has [Greek: thromboi] alone, without [Greek: haimatos]. Luke, [Greek: egeneto ho hidros autou hosei thromboi haimatos katabainontes]. Justin, [Greek: hidros hosei thromboi katecheito].

(2) That this is regarded as a fulfilment of Ps. xxii. 14 ('All my tears are poured out' &c.).

(3) That in continuing the quotation Justin follows Matthew rather than Luke. These considerations may be held to qualify, though I do not think that they suffice to remove, the conclusion that St. Luke's Gospel is being quoted. It seems to be sufficiently clear that [Greek: thromboi] might be used in this signification without [Greek: aimatos] [Endnote 124:2], and it appears from the whole manner of Justin's narrative that he intends to give merely the sense and not the words, with the exception of the single saying 'Let this cup pass from Me,' which is taken from St. Matthew. We cannot say positively that this feature did not occur in any other Gospel, but there is absolutely no reason apart from this passage to suppose that it did. The construction with [Greek: hosei] is in some degree characteristic of St. Luke, as it occurs more often in the works of that writer than in all the rest of the New Testament put together.

In narrating the institution of the Lord's Supper Justin has the clause which is found only in St. Luke and St. Paul, 'This do in remembrance of Me' ([Greek: mou] for [Greek: emaen]). The giving of the cup he quotes rather after the first two Synoptics, and adds 'that He gave it to them (the Apostles) alone.' This last does not seem to be more than an inference of Justin's own.

Two other sayings Justin has which are without parallel except in St. Luke. One is from the mission of the seventy.

Justin, Dial. 76

[Greek: Didomi humin exousian katapatein epano opheon, kai skorpion, kai skolopendron, kai epano parsaes dunameos tou echthrou.]

Luke x. 19.

[Greek: Idou, didomi humin taen exousian tou patein epano epheon, kai skorpion, kai epi pasan taen dunamin tou echthrou.]

The insertion of [Greek: skolopendron] here is curious. It may be perhaps to some extent paralleled by the insertion of [Greek: kai eis thaeran] in Rom. xi. 9: we have also seen a strange addition in the quotation of Ps. li. 19 in the Epistle of Barnabas (c. ii). Otherwise the resemblance of Justin to the Gospel is striking. The second saying, 'To whom God has given more, of him shall more be required' (Apol. i. 17), if quoted from the Gospel at all, is only a paraphrase of Luke xii. 48.

Besides these there are other passages, which are perhaps stronger as separate items of evidence, where, in quoting synoptic matter, Justin makes use of phrases which are found only in St. Luke and are discountenanced by the other Evangelists. Thus in the account of the rich young man, the three synoptical versions of the saying that impossibilities with men are possible with God, run thus:—

Luke xviii. 27.

[Greek: Ta adunata para anthropois dunata para to Theo estin.]

Mark x. 27.

[Greek: Para anthropois adunaton, all' ou para Theo; punta gar dunata para to Theo].

Matt. xix. 26.

[Greek: Para anthropois touto adunaton estin, para de Theo dunata panta].

Here it will be observed that Matthew and Mark (as frequently happens) are nearer to each other than either of them is to Luke. This would lead us to infer that, as they are two to one, they more nearly represent the common original, which has been somewhat modified in the hands of St. Luke. But now Justin has the words precisely as they stand in St. Luke, with the omission of [Greek: estin], the order of which varies in the MSS. of the Gospel. This must be taken as a strong proof that Justin has used the peculiar text of the third Gospel. Again, it is to be noticed that in another section of the triple synopsis (Mark xii. 20=Matt. xxii. 30=Luke xx. 35, 36) he has, in common with Luke and diverging from the other Gospels which are in near agreement, the remarkable compound [Greek: isangeloi] and the equally remarkable phrase [Greek: huioi taes anastaseos] ([Greek: tekna tou Theou taes anastaseos] Justin). This also I must regard as supplying a strong argument for the direct use of the Gospel. Many similar instances may be adduced; [Greek: erchetai] ([Greek: aexei] Justin) [Greek: ho ischuroteros] (Luke iii. 16), [Greek: ho nomos kai hoi prophaetai heos] ([Greek: mechri] Justin) [Greek: Ioannon] (Luke xvi. 16), [Greek: panti to aitounti] (Luke vi. 30), [Greek: to tuptonti se epi] ([Greek: sou] Justin) [Greek: taen siagona pareche kai taen allaen k.t.l.] (Luke vi. 29; compare Matt. v. 39, 40), [Greek: ti me legeis agathon] and [Greek: oudeis agathos ei mae] (Luke xviii. 19; compare Matt. xix. 17), [Greek: meta tauta mae echonton] ([Greek: dunamenous] Justin) [Greek: perissoteron] (om. Justin) [Greek: ti poiaesae k.t.l.] (Luke xii. 4, 5; compare Matt. X. 28), [Greek: paeganon] and [Greek: agapaen tou Theou] (Luke xi. 42). In the parallel passage to Luke ix. 22 (Matt xvi. 21 Mark viii. 31) Justin has the striking word [Greek: apodokimasthaenai], with Mark and Luke against Matthew, and [Greek: hupo] with Mark against the [Greek: apo] of the two other Synoptics. This last coincidence can perhaps hardly be pressed, as [Greek: hupo] would be the more natural word to use.

In the cases where we have only the double synopsis to compare with Justin, we have no certain test to distinguish between the primary and secondary features in the text of the Gospels. We cannot say with confidence what belonged to the original document and what to the later editor who reduced it to its present form. In these cases therefore it is possible that when Justin has a detail that is found in St. Matthew and wanting in St. Luke, or found in St. Luke and wanting in St. Matthew, he is still not quoting directly from either of those Gospels, but from the common document on which they are based. The triple synopsis however furnishes such a criterion. It enables us to see what was the original text and how any single Evangelist has diverged from it. Thus in the two instances quoted at the beginning of the last paragraph it is evident that the Lucan text represents a deviation from the original, and that deviation Justin has reproduced. The word [Greek: isangeloi] may be taken as a crucial case. Both the other Synoptics have simply [Greek hos angeloi], and this may be set down as undoubtedly the reading of the original; the form [Greek: isangeloi], which occurs nowhere else in the New Testament, and I believe, so far as we know, nowhere else in Greek before this passage [Endnote 128:1], has clearly been coined by the third Evangelist and has been adopted from him by Justin. So that in a quotation which otherwise presents considerable variation we have what I think must be called the strongest evidence that Justin really had St. Luke's narrative, either in itself or in some secondary shape, before him.

We are thus brought once more to the old result. If Justin did not use our Gospels in their present shape as they have come down to us, he used them in a later shape, not in an earlier. His resemblances to them cannot be accounted for by the supposition that he had access to the materials out of which they were composed, because he reproduces features which by the nature of the case cannot have been present in those originals, but of which we are still able to trace the authorship and the exact point of their insertion. Our Gospels form a secondary stage in the history of the text, Justin's quotations a tertiary. In order to reach the state in which it is found in Justin, the road lies through our Gospels, and not outside them.

This however does not exclude the possibility that Justin may at times quote from uncanonical Gospels as well. We have already seen reason to think that he did so from the substance of the Evangelical narrative, as it appears in his works, and this conclusion too is not otherwise than confirmed by its form. The degree and extent of the variations incline us to introduce such an additional factor to account for them. Either Justin has used a lost Gospel or Gospels, besides those that are still extant, or else he has used a recension of these Gospels with some slight changes of language and with some apocryphal additions. We have seen that he has two short sayings and several minute details that are not found in our present Gospels. A remarkable coincidence is noticed in 'Supernatural Religion' with the Protevangelium of James [Endnote 129:1]. As in that work so also in Justin, the explanation of the name Jesus occurs in the address of the angel to Mary, not to Joseph, 'Behold thou shalt conceive of the Holy Ghost and bear a Son and He shall be called the Son of the Highest, and thou shalt call His name Jesus, for He shall save His people from their sins.' Again the Protevangelium has the phrase 'Thou shalt conceive of His Word,' which, though not directly quoted, appears to receive countenance from Justin. The author adds that 'Justin's divergences from the Protevangelium prevent our supposing that in its present form it could have been the actual source of his quotations,' though he thinks that he had before him a still earlier work to which both the Protevangelium and the third Gospel were indebted. So far as the Protevangelium is concerned this may very probably have been the case; but what reason there is for assuming that the same document was also anterior to the third Gospel I am not aware. On the contrary, this very passage seems to suggest an opposite conclusion. The quotation in Justin and the address in the Protevangelium both present a combination of narratives that are kept separate in the first and third Gospels. But this very fact supplies a strong presumption that the version of those Gospels is the earliest. It is unlikely that the first Evangelist, if he had found his text already existing as part of the speech of the angel to Mary, would have transferred it to an address to Joseph; and it is little less unlikely that the third Evangelist, finding the fuller version of Justin and the Protevangelium, should have omitted from it one of its most important features. If a further link is necessary to connect Justin with the Protevangelium, that link comes into the chain after our Gospels and not before. Dr. Hilgenfeld has also noticed the phrase [Greek: charan de labousa Mariam] as common to Justin and the Protevangelium [Endnote 130:1]. This, too, may belong to the older original of the latter work. The other verbal coincidences with the Gospel according to the Hebrews in the account of the Baptism, and with that of Thomas in the 'ploughs and yokes,' have been already mentioned, and are, I believe, along with those just discussed, all that can be directly referred to an apocryphal source.

Besides these there are some coincidences in form between quotations as they appear in Justin and in other writers, such as especially the Clementine Homilies. These are thought to point to the existence of a common Gospel (now lost) from which they may have been extracted. It is unnecessary to repeat what has been said about one of these passages ('Let your yea be yea,' &c.). Another corresponds roughly to the verse Matt. xxv. 41, where both Justin and the Clementine Homilies read [Greek: hupagete eis to skotos to exoteron o haetoimasen ho pataer to satana (to diabolo] Clem. Hom.) [Greek: kai tois angelois autou] for the canonical [Greek: poreuesthe ap' emou eis to pur to aionion to haetoimasmenon k.t.l.] It is true that there is a considerable approximation to the reading of Justin and the Clementines, found especially in MSS. and authorities of a Western character (D. Latt. Iren. Cypr. Hil.), but there still remains the coincidence in regard to [Greek: exoteron](?) for [Greek: aionion] and [Greek: skotos] for [Greek: pyr], which seems to be due to something more than merely a variant text of the Gospel. A third meeting-point between Justin and the Clementines is afforded by a text which we shall have to touch upon when we come to speak of the fourth Gospel. Of the other quotations common to the Clementines and Justin there is a partial but not complete coincidence in regard to Matt. vii. 15, xi. 27, xix. 16, and Luke vi. 36. In Matt. vii. 15 the Clementines have [Greek: polloi eleusontai] where Justin has once [Greek: polloi eleusontai], once [Greek: polloi aexousin], and once the Matthaean version [Greek: prosechete apo ton pseudoprophaeton oitines erchontai k.t.l.] There is however a difference in regard to the reading [Greek: en endumasi], where the Clementines have [Greek: en endumatie], and Justin twice over [Greek: endedumenoi]. In Matt. xi. 27, Justin and the Clementines agree as to the order of the clauses, and twice in the use of the aorist [Greek: egno] (Justin has once [Greek: ginosko]), but in the concluding clause ([Greek: ho [ois] Clem.] [Greek: ean boulaetai ho nios apokalupsai]) Justin has uniformly in the three places where the verse is quoted [Greek: ois an ho uhios apokalupsae]. In Matt. xix. 16, 17 (Luke xviii. 18, 19) the Clementines and Justin alternately adhere to the Canonical text while differing from each other, but in the concluding phrase Justin has on one occasion the Clementine reading, [Greek: ho pataer mou ho en tois ouranois]. In Luke vi. 36 the Clementines have [Greek: ginesthe agathoi kai ioktirmones], where Justin has [Greek: ginesthe chraestoi kai oiktirmones] against the Canonical [Greek: ginesthe oiktirmones]. On the other hand, it should be said that the remaining quotations common to the Clementines and Justin have to all appearance no relation to each other. This applies to Matt. iv. 10, v. 39, 40, vi. 8, viii. 11, x. 28; Luke xi. 52. Speaking generally we seem to observe in comparing Justin and the Clementines phenomena not dissimilar to those which appear on a comparison with the Canonical Gospels. There is perhaps about the same degree at once of resemblance and divergence.

The principal textual coincidence with other writers is that with the Gospel used by the Marcosians as quoted by Irenaeus (Adv. Haer. i. 20. 3). Here the reading of Matt. xi. 27 is given in a form very similar to that of Justin, [Greek: oudeis hegno ton patera ei mae ho uhios, kai (oude Justin) ton uhion, ei mae ho pataer kai ho (ois] Justin) [Greek: an ho uhios apokalupsae]. This verse however is quoted by the early writers, orthodox as well as heretical, in almost every possible way, and it is not clear from the account in Irenaeus whether the Marcosians used an extra- canonical Gospel or merely a different text of the Canonical. Irenaeus himself seems to hold the latter view, and in favour of it may be urged the fact that they quote passages peculiar both to the first and the third Gospel; on the other hand, one of their quotations, [Greek: pollakis epethuaesa akousai hena ton logon touton], does not appear to have a canonical original.

On reviewing these results we find them present a chequered appearance. There are no traces of coincidence so definite and consistent as to justify us in laying the finger upon any particular extra-canonical Gospel as that used by Justin. But upon the whole it seems best to assume that some such Gospel was used, certainly not to the exclusion of the Canonical Gospels, but probably in addition to them.

A confusing element in the whole question is that to which we have just alluded in regard to the Gospel of the Marcosians. It is often difficult to decide whether a writer has really before him an unknown document or merely a variant text of one with which we are familiar. In the case of Justin it is to be noticed that there is often a very considerable approximation to his readings, not in the best text, but in some very early attested text, of the Canonical Gospels. It will be well to collect some of the most prominent instances of this.

Matt. iii. 15 ad fin. [Greek: kai pur anaephthae en to Iordanae] Justin. So a. (Codex Vercellensis of the Old Latin translation) adds 'et cum baptizaretur lumen ingens circumfulsit de aqua ita ut timerent onmes qui advenerant;' g[1]. (Codex Sangermanensis of the same) 'lumen magnum fulgebat de aqua,' &c. See above.

Luke iii. 22. Justin reads [Greek: uhios mon ei su, ego saemeron gegennaeka se]. So D, a, b, c, ff, l, Latin Fathers ('nonnulli codices' Augustine). See above.

Matt. v. 28. [Greek: hos un emblepsae] for [Greek: pas ho blepon]. Origen five times as Justin, only once the accepted text.

Matt. v. 29. Justin and Clement of Alexandria read here [Greek: ekkopson] for [Greek: exele], probably from the next verse or from Matt. xviii. 8.

Matt. vi. 20. [Greek: ouranois] Clem. Alex. with Justin; [Greek: ourano] the accepted reading.

Matt. xvi. 26. [Greek: opheleitai] Justin with most MSS. both of the Old Latin and of the Vulgate, the Curetonian Syriac (Crowfoot), Clement, Hilary, and Lucifer, against [Greek: ophelaethaesetai] of the best Alexandrine authorities.

Matt. vi. 21. There is a striking coincidence here with Clement of Alexandria, who reads, like Justin, [Greek: nous] for [Greek: cardia]; it would seem that Clement had probably derived his reading from Justin.

Matt. v. 22. [Greek: hostis an orgisthae] Syr. Crt. (Crowfoot); so Justin ([Greek: hos]).

Matt. v. 16. Clement of Alexandria (with Tertullian and several Latin Fathers) has [Greek: lampsato ta erga] and [Greek: ta agatha erga], where Justin has [Greek: lampsato ta kala erga], for [Greek: lampsato to phos]. Both readings would seem to be a gloss on the original.

Matt. v. 37. [Greek: kai] is inserted, as in Justin, by a, b, g, h, Syr. Crt. and Pst.

Luke x. 16. Justin has the reading [Greek: ho emou akouon akouei ton aposteilantos me]: so D, i, l (of the Old Latin) in place of [Greek: ho eme atheton k.t.l.]; in addition to it, E, a, b, Syr. Crt. and Hel. &c.

Matt. vii. 22. [Greek: ou to so anomati ephagomen kai epiomen] Justin; similarly Origen, four times, and Syr. Crt.

Luke xiii. 27. [Greek: anomias] for [Greek: adikias], D and Justin.

Matt. xiii. 43. [Greek: lampsosin] for [Greek: eklampsosin] with Justin, D, and Origen (twice).

Matt. xxv. 41. Of Justin's readings in this verse [Greek: hupagete] for [Greek: poreuesthe] is found also in [Hebrew: ?] and Hippolytus, [Greek: exoteron] for [Greek: aionion] in the cursive manuscript numbered 40 (Credner; I am unable to verify this), [Greek: ho haetoimasen ho pater mou] for [Greek: to haetoimasmenon] D. 1, most Codd. of the Old Latin, Iren. Tert. Cypr. Hil. Hipp. and Origen in the Latin translation.

Luke xii. 48. D, like Justin, has here [Greek: pleon] for [Greek: perissoteron] and also the compound form [Greek: apaitaesousin].

Luke xx. 24. Though in the main following (but loosely) the text of Luke, Justin has here [Greek: to nomisma], as Matt., instead of [Greek: daenarion]; so D.

Though it will be seen that Justin has thus much in common with D and the Old Latin version, it should be noticed that he has the verse, Luke xxii. 19, and especially the clause [Greek: touto poieite eis taen emaen anamnaesin] which is wanting in these authorities. On the other hand, he appears to have with them and other authorities, including Syr. Crt., the Agony in the Garden as given in Luke xxii, 43,44, which verses are omitted in MSS. of the best Alexandrine type. Luke xxiii. 34, Justin also has, with the divided support of the majority of Greek MSS. Vulgate, c, e, f, ff of the Old Latin, Syr. Crt. and Pst. &c. against B, D (prima manu), a, b, Memph. (MSS.) Theb.

These readings represent in the main a text which was undoubtedly current and widely diffused in the second century. 'Though no surviving manuscript of the Old Latin version dates before the fourth century and most of them belong to a still later age, yet the general correspondence of their text with that of the first Latin Fathers is a sufficient voucher for its high antiquity. The connexion subsisting between this Latin, version, the Curetonian Syriac and Codex Bezae, proves that the text of these documents is considerably older than the vellum on which they are written.' Such is Dr. Scrivener's verdict upon the class of authorities with which Justin shows the strongest affinity, and he goes on to add; 'Now it may be said without extravagance that no set of Scriptural records affords a text less probable in itself, less sustained by any rational principles of external evidence, than that of Cod. D, of the Latin codices, and (so far as it accords with them) of Cureton's Syriac. Interpolations as insipid in themselves as unsupported by other evidence abound in them all.... It is no less true to fact than paradoxical in sound, that the worst corruptions to which the New Testament has ever been subjected originated within a hundred years after it was composed' [Endnote 135:1]. This is a point on which text critics of all schools are substantially agreed. However much they may differ in other respects, no one of them has ever thought of taking the text of the Old Syriac and Old Latin translations as the basis of an edition. There can be no question that this text belongs to an advanced, though early, stage of corruption.

At the same stage of corruption, then, Justin's quotations from the Gospels are found, and this very fact is a proof of the antiquity of originals so corrupted. The coincidences are too many and too great all to be the result of accident or to be accounted for by the parallel influence of the lost Gospels. The presence, for instance, of the reading [Greek: o haetoimasen ho pataer] for [Greek: to haetoimasmenon] in Irenaeus and Tertullian (who has both 'quem praeparavit deus' and 'praeparatum') is a proof that it was found in the canonical text at a date little later than Justin's. And facts such as this, taken together with the arguments which make it little less than certain that Justin had either mediately or immediately access to our Gospels, render it highly probable that he had a form of the canonical text before him.

And yet large as is the approximation to Justin's text that may be made without stirring beyond the bounds of attested readings within the Canon, I still retain the opinion previously expressed that he did also make use of some extra-canonical book or books, though what the precise document was the data are far too insufficient to enable us to determine. So far as the history of our present Gospels is concerned, I have only to insist upon the alternative that Justin either used those Gospels themselves or else a later work, of the nature of a harmony based upon them [Endnote 136:1]. The theory (if it is really held) that he was ignorant of our Gospels in any shape, seems to me, in view of the facts, wholly untenable.



CHAPTER V.

HEGESIPPUS—PAPIAS.

Dr. Lightfoot has rendered a great service to criticism by his masterly exposure of the fallacies in the argument which has been drawn from the silence of Eusebius in respect to the use of the Canonical Gospels by the early writers [Endnote 138:1]. The author of 'Supernatural Religion' is not to be blamed for using this argument. In doing so he has only followed in the wake of the Germans who have handed it on from one to the other without putting it to a test so thorough and conclusive as that which has now been applied [Endnote 138:2]. For the future, I imagine, the question has been set at rest and will not need to be reopened [Endnote 138:3].

Dr. Lightfoot has shown, with admirable fulness and precision, that the object of Eusebius was only to note quotations in the case of books the admission of which into the Canon had been or was disputed. In the case of works, such as the four Gospels, that were universally acknowledged, he only records what seem to him interesting anecdotes or traditions respecting their authors or the circumstances under which they were composed. This distinction Dr. Lightfoot has established, not only by a careful examination of the language of Eusebius, but also by comparing his statements with the actual facts in regard to writings that are still extant, and where we are able to verify his procedure. After thus testing the references in Eusebius to Clement of Rome, the Ignatian Epistles, Polycarp, Justin, Theophilus of Antioch, and Irenaeus, Dr. Lightfoot arrives, by a strict and ample induction, at the conclusion that the silence of Eusebius in respect to quotations from any canonical book is so far an argument in its favour that it shows the book in question to have been generally acknowledged by the early Church. Instead of being a proof that the writer did not know the work in reference to which Eusebius is silent, the presumption is rather that he did, like the rest of the Church, receive it. Eusebius only records what seems to him specially memorable, except where the place of the work in or out of the Canon has itself to be vindicated.

But if this holds good, then most of what is said against the use of the Gospels by Hegesippus falls to the ground. Eusebius expressly says [Endnote 140:1] that Hegesippus made occasional use of the Gospel according to the Hebrews ([Greek: ek te tou kath' Hebraious euangeliou ... tina tithaesin]). But apart from the conclusion referred to above, the very language of Eusebius ([Greek: tithaesin tina ek]) is enough to suggest that the use of the Gospel according to the Hebrews was subordinate and subsidiary. Eusebius can hardly have spoken in this way of 'the Gospel of which Hegesippus made use' in all the five books of his 'Memoirs.' The expression tallies exactly with what we should expect of a work used in addition to but not to the exclusion of our Gospels. The fact that Eusebius says nothing about these shows that his readers would take it for granted that Hegesippus, as an orthodox Christian, received them.

With this conclusion the fragments of the work of Hegesippus that have come down to us agree. The quotations made in them are explained most simply and naturally, on the assumption that our Gospels have been used. The first to which we come is merely an allusion to the narrative of Matt. ii; 'For Domitian feared the coming of the Christ as much as Herod.' Those therefore who take the statement of Eusebius to mean that Hegesippus used only the Gospel according to the Hebrews are compelled to seek for the account of the Massacre of the Innocents in that Gospel. It appears however from Epiphanius that precisely this very portion of the first Gospel was wanting in the Gospel according to the Hebrews as used both by the Ebionites and by the Nazarenes. 'But if it be doubtful whether some forms of that Gospel contained the two opening chapters of Matthew, it is certain that Jerome found them in the version which he translated' [Endnote 141:1]. I am afraid that here, as in so many other cases, the words 'doubtful' and 'certain' are used with very little regard to their meanings. In support of the inference from Jerome, the author refers to De Wette, Schwegler, and an article in a periodical publication by Ewald. De Wette expressly says that the inference does not follow ('Aus Comm. ad Matt. ii. 6 ... laesst sich nicht schliessen dass er hierbei das Evang. der Hebr. verglichen habe.... Nicht viel besser beweisen die St. ad Jes. xi. 1; ad Abac. iii. 3') [Endnote 141:2]. He thinks that the presence of these chapters in Jerome's copy cannot be satisfactorily proved, but is probable just from this allusion in Hegesippus—in regard to which De Wette simply follows the traditional, but, as we have seen, erroneous assumption that Hegesippus used only the Gospel according to the Hebrews. Schwegler [Endnote 141:3] gives no reasons, but refers to the passages quoted from Jerome in Credner. Credner, after examining these passages, comes to the conclusion that 'the Gospel of the Nazarenes did not contain the chapters' [Endnote 141:4]. Ewald's periodical I cannot refer to, but Hilgenfeld, after an elaborate review of the question, decides that the chapters were omitted [Endnote 141:5]. This is the only authority I can find for the 'certainty that Jerome found them' in his version.

On the whole, then, it seems decidedly more probable (certainties we cannot deal in) that the incident referred to by Hegesippus was missing from the Gospel according to the Hebrews. That Gospel therefore was not quoted by him, but, on the contrary, there is a presumption that he is quoting from the Canonical Gospel. The narrative of the parallel Gospel of St. Luke seems, if not to exclude the Massacre of the Innocents, yet to imply an ignorance of it.

The next passage that appears to be quotation occurs in the account of the death of James the Just; 'Why do ye ask me concerning Jesus the Son of Man? He too sits in heaven on the right hand of the great Power and will come on the clouds of heaven' ([Greek: Ti me eperotate peri Iaesou tou huiou tou anthropou? kai autos kathaetai en to ourano ek dexion taes megalaes dunameos, kai mellei erchesthai epi ton nephelon tou ouranou]). It seems natural to suppose that this is an allusion to Matt. xxvi. 64, [Greek: ap' arti opsesthe ton huion tou anthropou kathaemenon ek dexion taes dunameos, kai erchomenon epi ton vephelon tou ouranou]. The passage is one that belongs to the triple synopsis, and the form in which it appears in Hegesippus shows a preponderating resemblance to the version of St. Matthew. Mark inserts [Greek: kathaemenon] between [Greek: ek dexion] and [Greek: taes dunameos], while Luke thinks it necessary to add [Greek: tou theou]. The third Evangelist omits the phrase [Greek: epi ton nephelon tou ouranou], altogether, and the second substitutes [Greek: meta] for [Greek: epi]. In fact the phrase [Greek: epi ton vephelon] occurs in the New Testament only in St. Matthew; the Apocalypse, like St. Mark, has [Greek: meta] and [Greek: epi] only with the singular.

In like manner, when we find Hegesippus using the phrase [Greek: prosopon ou lambaneis], this seems to be a reminiscence of Luke xx. 21, where the synoptic parallels have [Greek: blepeis].

A more decided reference to the third Gospel occurs in the dying prayer of St. James; [Greek: parakalo, kurie thee pater, aphes autois; ou gar oidasiti poiousin], which corresponds to Luke xxiii. 34, [Greek: pater, aphes autois; ou gar oidasin ti poiousin]. There is the more reason to believe that Hegesippus' quotation is derived from this source that it reproduces the peculiar use of [Greek: aphienai] in the sense of 'forgive' without an expressed object. Though the word is of very frequent occurrence, I find no other instance of this in the New Testament [Endnote 143:1], and the Clementine Homilies, in making the same quotation, insert [Greek: tas hamartias auton]. The saying is well known to be peculiar to St. Luke. There is perhaps a balance of evidence against its genuineness, but this is of little importance, as it undoubtedly formed part of the Gospel as early as Irenaeus, who wrote much about the same time as Hegesippus.

The remaining passage occurs in a fragment preserved from Stephanus Gobarus, a writer of the sixth century, by Photius, writing in the ninth. Referring to the saying 'Eye hath not seen,' &c., Gobarus says 'that Hegesippus, an ancient and apostolical man, asserts—he knows not why—that these words are vainly spoken, and that those who use them give the lie to the sacred writings and to our Lord Himself who said, "Blessed are your eyes that see and your ears that hear,"' &c. 'Those who use these words' are, we can hardly doubt, as Dr. Lightfoot after Routh has shown [Endnote 144:1], the Gnostics, though Hegesippus would seem to have forgotten I Cor. ii. 9. The anti-Pauline position assigned to Hegesippus on the strength of this is, we must say, untenable. But for the present we are concerned rather with the second quotation, which agrees closely with Matt. xiii. 26 ([Greek: humon de makarioi hoi ophthalmoi hoti blepousin, kai ta ota humon hoti akouousin]). The form of the quotation has a slightly nearer resemblance to Luke x. 23 ([Greek: makarioi hoi ophthalmoi hoi blepontes ha blepete k.t.l.]), but the marked difference in the remainder of the Lucan passage increases the presumption that Hegesippus is quoting from the first Gospel [Endnote 144:2].

The use of the phrase [Greek: ton theion graphon] is important and remarkable. There is not, so far as I am aware, any instance of so definite an expression being applied to an apocryphal Gospel. It would tend to prepare us for the strong assertion of the Canon of the Gospels in Irenaeus; it would in fact mark the gradually culminating process which went on in the interval which separated Irenaeus from Justin. To this interval the evidence of Hegesippus must be taken to apply, because though writing like Irenaeus under Eleutherus (from 177 A.D.) he was his elder contemporary, and had been received with high respect in Rome as early as the episcopate of Anicetus (157-168 A.D.).

The relations in which Hegesippus describes himself as standing to the Churches and bishops of Corinth and Rome seem to be decisive as to his substantial orthodoxy. This would give reason to think that he made use of our present Gospels, and the few quotations that have come down to us confirm that view not inconsiderably, though by themselves they might not be quite sufficient to prove it.

There is one passage that may be thought to point to an apocryphal Gospel, 'From these arose false Christs, false prophets, false apostles;' which recalls a sentence in the Clementines, 'For there shall be, as the Lord said, false apostles, false prophets, heresies, ambitions.' It is not, however, nearer to this than to the canonical parallel, Matt. xxiv. 24 ('There shall arise false Christs and false prophets').

2.

In turning from Hegesippus to Papias we come at last to what seems to be a definite and satisfactory statement as to the origin of two at least of the Synoptic Gospels, and to what is really the most enigmatic and tantalizing of all the patristic utterances.

Like Hegesippus, Papias may be described as 'an ancient and apostolic man,' and appears to have better deserved the title. He is said to have suffered martyrdom under M. Aurelius about the same time as Polycarp, 165-167 A.D. [Endnote 145:1] He wrote a commentary on the Discourses or more properly Oracles of the Lord, from which Eusebius extracted what seemed to him 'memorable' statements respecting the origin of the first and second Gospels. 'Matthew,' Papias said [Endnote 146:1], 'wrote the oracles ([Greek: ta logia]) in the Hebrew tongue, and every one interpreted them as he was able.' 'Mark, as the interpreter of Peter, wrote down accurately, though not in order, all that he remembered that was said or done by Christ. For he neither heard the Lord nor attended upon Him, but later, as I said, upon Peter, who taught according to the occasion and not as composing a connected narrative of the Lord's discourses; so that Mark made no mistake in writing down some things as he remembered them. For he took care of one thing, not to omit any of the particulars that he heard or to falsify any part of them.'

* * * * *

Let us take the second of these statements first. According to it the Gospel of St. Mark consisted of notes taken down, or rather recollected, from the teaching of Peter. It was not written 'in order,' but it was an original work in the sense that it was first put in writing by Mark himself, having previously existed only in an oral form.

Does this agree with the facts of the Gospel as it appears to us now? There is a certain ambiguity as to the phrase 'in order.' We cannot be quite sure what Papias meant by it, but the most natural conclusion seems to be that it meant chronological order. If so, the statement of Papias seems to be so far borne out that none of the Synoptic Gospels is really in exact chronological order; but, strange to say, if there is any in which an approach to such an order is made, it is precisely this of St. Mark. This appears from a comparison of the three Synoptics. From the point at which the second Gospel begins, or, in other words, from the Baptism to the Crucifixion, it seems to give the outline that the other two Gospels follow [Endnote 147:1]. If either of them diverges from it for a time it is only to return. The early part of St. Matthew is broken up by the intrusion of the so-called Sermon on the Mount, but all this time St. Mark is in approximate agreement with St. Luke. For a short space the three Gospels go together. Then comes a second break, where Luke introduces his version of the Sermon on the Mount. Then the three rejoin and proceed together, Matthew being thrown out by the way in which he has collected the parables into a single chapter, and Luke later by the place which he has assigned to the incident at Nazareth. After this Matthew and Mark proceed side by side, Luke dropping out of the ranks. At the confession of Peter he takes his place again, and there is a close agreement in the order of the three narratives. The incident of the miracle-worker is omitted by Matthew, and then comes the insertion of a mass of extraneous matter by Luke. When he resumes the thread of the common narrative again all three are together. The insertion of a single parable on the part of Matthew, and omissions on the part of Luke, are the only interruptions. There is an approximate agreement of all three, we may say, for the rest of the narrative. We observe throughout that, in by far the preponderating number of instances, where Matthew differs from the order of Mark, Luke and Mark agree, and where Luke differs from the order of Mark, Matthew and Mark agree. Thus, for instance, in the account of the healings in Peter's house and of the paralytic, in the relation of the parables of Mark iv. 1-34 to the storm at sea which follows, of the healing of Jairus' daughter to that of the Gadarene demoniac and to the mission of the Twelve in the place of Herod's reflections (Mark vi. 14-16), in the warning against the Scribes and the widow's mite (Mark xii. 38-44), the second and third Synoptics are allied against the first. On the other hand, in the call of the four chief Apostles, the death of the Baptist, the walking on the sea, the miracles in the land of Gennesareth, the washing of hands, the Canaanitish woman, the feeding of the four thousand and the discourses which follow, the ambition of the sons of Zebedee, the anointing at Bethany, and several insertions of the third Evangelist in regard to the last events, the first two are allied against him. While Mark thus receives such alternating support from one or other of his fellow Evangelists, I am not aware of any clear case in which, as to the order of the narratives, they are, united and he is alone, unless we are to reckon as such his insertion of the incident of the fugitive between Matt. xxvi. 56, 57, Luke xxii. 53, 54.

It appears then that, so far as there is an order in the Synoptic Gospels, the normal type of that order is to be found precisely in St. Mark, whom Papias alleges to have written not in order.

But again there seems to be evidence that the Gospel, in the form in which it has come down to us, is not original but based upon another document previously existing. When we come to examine closely its verbal relations to the other two Synoptics, its normal character is in the main borne out, but still not quite completely. The number of particulars in which Matthew and Mark agree together against Luke, or Mark and Luke agree together against Matthew, is far in excess of that in which Matthew and Luke are agreed against Mark. Mark is in most cases the middle term which unites the other two. But still there remains a not inconsiderable residuum of cases in which Matthew and Luke are in combination and Mark at variance. The figures obtained by a not quite exact and yet somewhat elaborate computation [Endnote 149:1] are these; Matthew and Mark agree together against Luke in 1684 particulars, Luke and Mark against Matthew in 944, but Matthew and Luke against Mark in only 334. These 334 instances are distributed pretty evenly over the whole of the narrative. Thus (to take a case at random) in the parallel narratives Matt. xii. 1-8, Mark ii. 23-28, Luke vi. 1-5 (the plucking of the ears on the Sabbath day), there are fifty-one points (words or parts of words) common to all three Evangelists, twenty-three are common only to Mark and Luke, ten to Mark and Matthew, and eight to Matthew and Luke. In the next section, the healing of the withered hand, twenty points are found alike in all three Gospels, twenty-seven in Mark and Luke, twenty-one in Mark and Matthew, and five in Matthew and Luke. Many of these coincidences between the first and third Synoptics are insignificant in the extreme. Thus, in the last section referred to (Mark iii. 1-6=Matt. xii. 9-14=Luke vi. 6-11), one is the insertion of the article [Greek: taen] ([Greek: sunagogaen]), one the insertion of [Greek: sou] ([Greek: taen cheira sou]), two the use of [Greek: de] for [Greek: kai], and one that of [Greek: eipen] for [Greek: legei]. In the paragraph before, the eight points of coincidence between Matthew and Luke are made up thus, two [Greek: kai aesthion] (=[Greek kai esthiein]), [Greek: eipon] (=[Greek: eipan]), [Greek: poiein, eipen, met' autou] (=[Greek: sun auto]), [Greek: monous] (=[Greek: monois]). But though such points as these, if they had been few in number, might have been passed without notice, still, on the whole, they reach a considerable aggregate and all are not equally unimportant. Thus, in the account of the healing of the paralytic, such phrases is [Greek epi klinaes, apaelthen eis ton oikon autou], can hardly have come into the first and third Gospels and be absent from the second by accident; so again the clause [Greek: alla ballousin (blaeteon) oinon neon eis askous kainous]. In the account of the healing of the bloody flux the important word [Greek: tou kraspedou] is inserted in Matthew and Luke but not in Mark; in that of the mission of the twelve Apostles, the two Evangelists have, and the single one has not, the phrase [Greek: kai therapeuein noson (nosous]), and the still more important clause [Greek: lego humin anektoteron estai (gae) Sodomon ... en haemera ... ae tae polei ekeinae]: in Luke ix. 7 (= Matt. xiv. 1) Herod's title is [Greek: tetrarchaes], in Mark vi. 14 [Greek: basileus]; in the succeeding paragraph [Greek: hoi ochloi aekolouthaesan] and the important [Greek: to perisseuon (-san)] are wanting in the intermediate Gospel; in the first prophecy of the Passion it has [Greek: apo] where the other two have [Greek: hupo], and [Greek: meta treis haemeras] where they have [Greek: tae tritae haemera]: in the healing of the lunatic boy it omits the noticeable [Greek: kai diestrammenae]: in the second prophecy of the Passion it omits [Greek: mellei], in the paragraph about offences, [Greek: elthein ta skandala ...ouai...di hou erchetai]. These points might be easily multiplied as we go on; suffice it to say that in the aggregate they seem to prove that the second Gospel, in spite of its superior originality and adhesion to the normal type, still does not entirely adhere to it or maintain its primary character throughout. The theory that we have in the second Gospel one of the primitive Synoptic documents is not tenable.

No doubt this is an embarrassing result. The question is easy to ask and difficult to answer—If our St. Mark does not represent the original form of the document, what does represent it? The original document, if not quite like our Mark, must have been very nearly like it; but how did any writer come to reproduce a previous work with so little variation? If he had simply copied or reproduced it without change, that would have been intelligible; if he had added freely to it, that also would have been intelligible: but, as it is, he seems to have put in a touch here and made an erasure there on principles that it is difficult for us now to follow. We are indeed here at the very crux of Synoptic criticism.

For our present purpose however it is not necessary that the question should be solved. We have already obtained an answer on the two points raised by Papias. The second Gospel is written in order; it is not an original document. These two characteristics make it improbable that it is in its present shape the document to which Papias alludes.

Does his statement accord any better with the phenomena of the first Gospel? He asserts that it was originally written in Hebrew, and that the large majority of modern critics deny to have been the case with our present Gospel. Many of the quotations in it from the Old Testament are made directly from the Septuagint and not from the Hebrew. There are turns of language which have the stamp of an original Greek idiom and could not have come in through translation. But, without going into this question as to the original language of the first Gospel, a shorter method will be to ask whether it can have been an original document at all? The work to which Papias referred clearly was such, but the very same investigation which shows that our present St. Mark was not original, tells with increased force against St. Matthew. When a document exists dealing with the same subject-matter as two other documents, and those two other documents agree together and differ from it on as many as 944 separate points, there can be little doubt that in the great majority of those points it has deviated from the original, and that it is therefore secondary in character. It is both secondary and secondary on a lower stage than St. Mark: it has preserved the features of the original with a less amount of accuracy. The points of the triple synopsis on which Matthew fails to receive verification are in all 944; those on which Mark fails to receive verification 334; or, in other words, the inaccuracies of Matthew are to those of Mark nearly as three to one. In the case of Luke the proportion is still greater— as much as five to one.

This is but a tithe of the arguments which show that the first Gospel is a secondary composition. An original composition would be homogeneous; it is markedly heterogeneous. The first two chapters clearly belong to a different stock of materials from the rest of the Gospel. A broad division is seen in regard to the Old Testament quotations. Those which are common to the other two Synoptists are almost if not quite uniformly taken from the Septuagint; those, on the other hand, which seem to belong to the reflection of the Evangelist betray more or less distinctly the influence of the Hebrew [Endnote 153:1]. Our Gospel is thus seen to be a recension of another original document or documents and not an original document itself.

Again, if our St. Matthew had been an original composition and had appeared from the first in its present full and complete form, it would be highly difficult to account for the omissions and variations in Mark and Luke. We should be driven back, indeed, upon all the impossibilities of the 'Benutzungs-hypothese.' On the one hand, the close resemblance between the three compels us to assume that the authors have either used each other's works or common documents; but the differences practically preclude the supposition that the later writer had before him the whole work of his predecessor. If Luke had had before him the first two chapters of Matthew he could not have written his own first two chapters as he has done.

Again, the character of the narrative is such as to be inconsistent with the view that it proceeds from an eye-witness of the events. Those graphic touches, which are so conspicuous in the fourth Gospel, and come out from time to time in the second, are entirely wanting in the first. If parallel narratives, such as the healing of the paralytic, the cleansing of the Temple, or the feeding of the five thousand, are compared, this will be very clearly seen. More; there are features in the first Gospel that are to all appearance unhistorical and due to the peculiar method of the writer. He has a way of reduplicating, so to speak, the personages of one narrative in order to make up for the omission of another [Endnote 154:1]. For instance, he is silent as to the healing of the demoniac at Capernaum, but, instead of this, he gives us two Gadarene demoniacs, at the same time modifying the language in which he describes this latter incident after the pattern of the former; in like manner he speaks of the healing of two blind men at Jericho, but only because he had passed over the healing of the blind man at Bethsaida. Of a somewhat similar nature is the adding of the ass's colt to the ass in the account of the Triumphal Entry. There are also fragmentary sayings repeated in the Gospel in a way that would be natural in a later editor piecing together different documents and finding the same saying in each, but unnatural in an eye- and ear-witness drawing upon his own recollections. Some clear cases of this kind would be Matt. v. 29, 30 (= Matt. xviii. 8, 9) the offending member, Matt. v. 32 (= Matt. xix. 9) divorce, Matt. x. 38, 39 (= Matt. xvi. 24, 25) bearing the cross, loss and gain; and there are various others.

These characteristics of the first Gospel forbid us to suppose that it came fresh from the hands of the Apostle in the shape in which we now have it; they also forbid us to identify it with the work alluded to by Papias. Neither of the two first Gospels, as we have them, complies with the conditions of Papias' description to such an extent that we can claim Papias as a witness to them.

* * * * *

But now a further enquiry opens out upon us. The language of Papias does not apply to our present Gospels; will it apply to some earlier and more primary state of those Gospels, to documents incorporated in the works that have come down to us but not co-extensive with them? German critics, it is well known, distinguish between 'Matthaeus'—the present Gospel that bears the name of St. Matthew—and 'Ur-Matthaeus,' or the original work of that Apostle, 'Marcus'—our present St. Mark—and 'Ur-Marcus,' an older and more original document, the real production of the companion of St. Peter. Is it to these that Papias alludes?

Here we have a much more tenable and probable hypothesis. Papias says that Matthew composed 'the oracles' ([Greek: ta logia]) in the Hebrew tongue. The meaning of the word [Greek: logia] has been much debated. Perhaps the strictest translation of it is that which has been given, 'oracles'—short but weighty and solemn or sacred sayings. I should be sorry to say that the word would not bear the sense assigned to it by Dr. Westcott, who paraphrases it felicitously (from his point of view) by our word 'Gospel' [Endnote 155:1]. It is, however, difficult to help feeling that the natural sense of the word has to be somewhat strained in order to make it cover the whole of our present Gospel, and to bring under it the record of facts to as great an extent as discourse. It seems at least the simplest and most obvious interpretation to confine the word strictly or mainly to discourse. 'Matthew composed the discourses (those brief yet authoritative discourses) in Hebrew.'

At this point we are met by a further coincidence. The common matter in the first three Gospels is divided into a triple synopsis and a double synopsis—the first of course running through all three Gospels, the second found only in St. Matthew and St. Luke. But this double synopsis is nearly, though not quite, confined to discourse; where it contains narration proper, as in the account of John the Baptist and the Centurion of Capernaum, discourse is largely mingled with it. But, if the matter common to Matthew and Luke consists of discourse, may it not be these very [Greek: logia] that Papias speaks of? Is it not possible that the two Evangelists had access to the original work of St. Matthew and incorporated its material into their own Gospels in different ways? It would thus be easy to understand how the name that belonged to a special and important part of the first Gospel gradually came to be extended over the whole. Bulk would not unnaturally be a great consideration with the early Christians. The larger work would quickly displace the smaller; it would contain all that the smaller contained with additions no less valuable, and would therefore be eagerly sought by the converts, whose object would be rather fulness of information than the best historical attestation. The original work would be simply lost, absorbed, in the larger works that grew out of it.

This is the kind of presumption that we have for identifying the Logia of Papias with the second ground document of the first Gospel—the document, that is, which forms the basis of the double synopsis between the first Gospel and the third. As a hypothesis the identification of these two documents seems to clear up several points. It gives a 'local habitation and a name' to a document, the separate and independent existence of which there is strong reason to suspect, and it explains how the name of St. Matthew came to be placed at the head of the Gospel without involving too great a breach in the continuity of the tradition. It should be remembered that Papias is not giving his own statement but that of the Presbyter John, which dates back to a time contemporary with the composition of the Gospel. On the other hand, by the time of Irenaeus, whose early life ran parallel with the closing years of Papias, the title was undoubtedly given to the Gospel in its present form. It is therefore as difficult to think that the Gospel had no connection with the Apostle whose name it bears, as it is impossible to regard it as entirely his work. The Logia hypothesis seems to suggest precisely such an intermediate relation as will satisfy both sides of the problem.

There are, however, still difficulties in the way. When we attempt to reconstruct the 'collection of discourses' the task is very far from being an easy one. We do indeed find certain groups of discourse in the first Gospel—such as the Sermon on the Mount ch. v-vii, the commission of the Apostles ch. x, a series of parables ch. xiii, of instructions in ch. xviii, invectives against the Pharisees in ch. xxvi, and long eschatological discourses in ch. xxiv and xxv, which seem at once to give a handle to the theory that the Evangelist has incorporated a work consisting specially of discourses into the main body of the Synoptic narrative. But the appearance of roundness and completeness which these discourses present is deceptive. If we are to suppose that the form in which the discourses appear in St. Matthew at all nearly represents their original structure, then how is it that the same discourses are found in the third Gospel in such a state of dispersion? How is it, for instance, that the parallel passages to the Sermon on the Mount are found in St. Luke scattered over chapters vi, xi, xii, xiii, xiv, xvi, with almost every possible inversion and variety of order? Again, if the Matthaean sections represent a substantive work, how are we to account for the strange intrusion of the triple synopsis into the double? What are we to say to the elaborately broken structure of ch. x? On the other hand, if we are to take the Lucan form as nearer to the original, that original must have been a singular agglomeration of fragments which it is difficult to piece together. It is easy to state a theory that shall look plausible so long as it is confined to general terms, but when it comes to be worked out in detail it will seem to be more and more difficult and involved at every step. The Logia hypothesis in fact carries us at once into the very nodus of Synoptic criticism, and, in the present state of the question, must be regarded as still some way from being established.

The problem in regard to St. Mark and the triple synopsis is considerably simpler. Here the difficulty arises from the necessity of assuming a distinction between our present second Gospel and the original document on which that Gospel is based. I have already touched upon this point. The synoptical analysis seems to conduct us to a ground document greatly resembling our present St. Mark, which cannot however be quite identical with it, as the Canonical Gospel is found to contain secondary features. But apart from the fact that these secondary features are so comparatively few that it is difficult to realise the existence of a work in which they, and they only, should be absent, there is this further obstacle to the identification even of the ground document with the Mark of Papias, that even in that original shape the Gospel still presented the normal type of the Synoptic order, though 'order' is precisely the characteristic that Papias says was, in this Gospel, wanting.

Everywhere we meet with difficulties and complexities. The testimony of Papias remains an enigma that can only be solved—if ever it is solved—by close and detailed investigations. I am bound in candour to say that, so far as I can see myself at present, I am inclined to agree with the author of 'Supernatural Religion' against his critics [Endnote 159:1], that the works to which Papias alludes cannot be our present Gospels in their present form.

What amount of significance this may have for the enquiry before us is a further question. Papias is repeating what he had heard from the Presbyter John, which would seem to take us up to the very fountainhead of evangelical composition. But such a statement does not preclude the possibility of subsequent changes in the documents to which it refers. The difficulties and restrictions of local communication must have made it hard for an individual to trace all the phases of literary activity in a society so widely spread as the Christian, even if it had come within the purpose of the writer or his informant to state the whole, and not merely the essential part, of what he knew.



CHAPTER VI.

THE CLEMENTINE HOMILIES.

It is unfortunate that there are not sufficient materials for determining the date of the Clementine Homilies. Once given the date and a conclusion of considerable certainty could be drawn from them; but the date is uncertain, and with it the extent to which they can be used as evidence either on one side or on the other.

Some time in the second century there sprang up a crop of heretical writings in the Ebionite sect which were falsely attributed to Clement of Rome. The two principal forms in which these have come down to us are the so-called Homilies and Recognitions. The Recognitions however are only extant in a Latin translation by Rufinus, in which the quotations from the Gospels have evidently been assimilated to the Canonical text which Rufinus himself used. They are not, therefore, in any case available for our purpose. Whether the Recognitions or the Homilies came first in order of time is a question much debated among critics, and the even way in which the best opinions seem to be divided is a proof of the uncertainty of the data. On the one side are ranged Credner, Ewald, Reuss, Schwegler, Schliemann, Uhlhorn, Dorner, and Luecke, who assign the priority to the Homilies: on the other, Hilgenfeld, Koestlin, Ritschl (doubtfully), and Volkmar, who give the first place to the Recognitions [Endnote 162:1]. On the ground of authority perhaps the preference should be given to the first of these, as representing more varied parties and as carrying with them the greater weight of sound judgment, but it is impossible to say that the evidence on either side is decisive.

The majority of critics assign the Clementines, in one form or the other, to the middle of the second century. Credner, Schliemann, Scholten, and Renan give this date to the Homilies; Volkmar and Hilgenfeld to the Recognitions; Ritschl to both recensions alike [Endnote 162:2]. We shall assume hypothetically that the Homilies are rightly thus dated. I incline myself to think that this is more probable, but, speaking objectively, the probability could not have a higher value put upon it than, say, two in three.

One reason for assigning the Homilies to the middle of the second century is presented by the phenomena of the quotations from the Gospels which correspond generally to those that are found in writings of this date, and especially, as has been frequently noticed, to those which we meet with in Justin. I proceed to give a tabulated list of the quotations. In order to bring out a point of importance I have indicated by a letter in the left margin the presence in the Clementine quotations of some of the peculiarities of our present Gospels. When this letter is unbracketed, it denotes that the passage is only found in the Gospel so indicated; when the letter is enclosed in brackets, it is implied that the passage is synoptical, but that the Clementines reproduce expressions peculiar to that particular Gospel. The direct quotations are marked by the letter Q. Many of the references are merely allusive, and in more it is sufficiently evident that the writer has allowed himself considerable freedom [Endnote 163:1].

Exact. Slightly variant. Variant. Remarks. (M.) 8.21, Luke 4.6-8 narrative. (Matt. 4.8-10), Q. 3.55, [Greek: ho ponaeros estin ho peirazon.], Q. 15.10, Matt. 5.3; Luke 6.20. M. 17.7, Matt. 5.8. (M.) 3.51 } Matt. 5. repeated Ep. Pet. 2} 17,18. identically. 11.32, Matt. 5. highly condensed 21-48. paraphrase, [Greek: oi en planae.] { Matt.5.44, allusive merely. 12.32 { 45(Luke 3.19 {6.27, 28, {35). M. 3.56, Matt. 5.34, 35, Q. M. 3.55} Matt. 5.37. repeated identi- 19.2} Q. cally; so Justin. (M.) 3.57. Matt. 5.45. Q. { oblique and allu- 12.26 { sive, repeated 18.2. { in part simi- 11.12 { larly; [Greek: { pherei ton { hueton]. M. 3.55, Matt. 6,6, Q. 19.2, Matt.6.13 Q. (M.) 3.55, Matt. 6.32; combination. 6.8 (Luke 12.30.) 18.16, Matt. 7.2 oblique and allu- (12). sive. 3.52, Matt. 7.7 [Greek: euris- (Luke 11.9). kete] for [euraeskete] in both. (L.M.) 3.56, Matt. 7.9-11 striking divi- (Luke 11.11-13) sion of pecu- liarities of both Gospels. 12.32} Matt. 7.12 repeated di- 7.4 } (Luke versely, 11.4 } 6.31. allusive. (M.) 18.17, Matt. 7. (omissions), Q. 13,14. 7.7. Matt. 7.13, allusive para- 14. phrase. (L.) 8.7, Luke 6.46. 11.35, Matt. 7.15. Justin, in part similarly, in part diversely. (M.) 8.4, Matt. 8.11, (addition), Q. Justin diversely. 12 (Luke 13.29). 9.21, Matt. 8.9 allusive merely. (Luke 7.8). (M.) 3.56, Matt. 9.13 (addition), Q. from LXX. (12.7). (L.M.) {Matt. 10. { { 13, 15= { { Luke 10. { 13.30, { 5,6,10- {mixed pecu- 31. { 12 (9.5) { liarities, { =Mark { oblique and { 6.11. { allusive. (L.M.) 17.5, Matt. 10.28 mixed peculia- (Luke 12,4, 5), Q. rities; Justin diversely. 12.31, Matt. 10. allusive merely. 29, 30 (Luke 12.6, 7). 3.17 {Matt. 11.11. allusive. {Luke 7.28. 8.6, Matt. 11.25 (addition)+. perhaps from (Luke x.21). Matt. 21.16. (M.) 17.4 } { 18.4 }Matt. 11.27 {repeated simi- 18.7 } (Luke { larly; cp. 18.13} 10.22), Q. { Justin, &c. 18.20} M. 3.52, Matt. (M.) +19.2. Matt. 12. [Greek: allae 26, Q. pou.] (M.L.) +19.7, Matt. 12. 34 (Luke 6. 45), Q. M.11.33, Matt. (addition), Q. 12.42. 11.33, Matt. 12. 41 (Luke 11. 32), Q. (M.L.) M.53, Matt. 13. 16 (Luke 10. 24), +Q. M.18.15, Matt. 13.35+. Mk. 19.20, Mark 4.34. M. 19.2, Matt. 13. 39, Q. M.3.52, Matt. 15. 15 (om. [Greek: mou]), Q. {Matt. 15. narrative. 11.19 {21-28 {(Mark [Greek: Iousta {7.24-30). Surophoini- kissa.] (M.) 17.18, Matt. 16. 16 (par.) M. Ep. Clem. 2, allusive merely. Matt. 16.19. M. Ep. Clem. 6, Matt. ditto. 16.19. (M.) 3.53, Matt. 17.5 (par.), Q. M. 12.29, Matt. 18. addition [Greek: 7, Q. ta agatha elthein.] M. 17.7, Matt. 18.10 (v.l.) (L.) 3.71, Luke 10.7. (order) (Matt.10.10). L. +19.2, Luke 10.18. L. 9.22, Luke 10.20. allusive merely. L. 17.5, Luke 18.6- 8, Q. (?) 19.2, [Greek: mae Cp. Eph. 4.27. dote prophasin to ponaero], Q. 3.53, Prophet like Cp. Acts 3.22. Moses, Q. (M.) 3.54, Matt. 19.8, sense more diver- 4 (Mark 10.5, gent than 6), Q. words. {Matt. 19. } 17.4 { 16,17. } 18.1 {Mark 10. }repeated simi- 18.3 { 17,18. } larly; cp. 18.17 {Luke 18. } Justin. 3.57 { 18,19. } L. 3.63, Luke 19. not quotation. 5.9. M.8.4, Matt. 22. 14, Q. (M.) 8.22, Matt. 22.9. allusive merely. 11. 3.50 {Matt. 22. } 2.51 {29 (Mark }repeated simi- 18.20 {12.24), Q. } larly. 3.50, [Greek: dia ti ou eulogon ton graphon;] (Mk.) 3.55, Mark 12.27 (par.), Mk. 3.57, Mark 12.29 [Greek: haemon], Q. 17.7, Mark 12.30 allusive. (Matt. 22.37). { 3.18, Matt. 23.2, M. { 3, Q. { 3.18, Matt. 23.13 repeated simi- { (Luke 11.52). larly. 18.15. (M.) 11.29, Matt. 23. 25, 26, Q. (Mk.) { 3.15, Mark 13.2 { (par.), Q. { 3.15, Matt. 24.3 { (par.), Q. L. { Luke 19.43, Q. 16.21, [Greek: esontai pseud- apostoloi]. (M.) 3.60 (3.64), Matt. part repeated 24.45-51 ( larly. Luke 12.42-46). (M.) 3.65, Matt. 25.21 (= Luke 19.17). (M. L.) 3.61, Matt. 25.26, ? mixed peculi- 26,27 (=Luke 19. arities. 22,23). 2.51}[Greek: 3.50} ginesthe 18.20} trapezitai } dokimoi.] M. 19.2. Matt. 25. [Greek: allae 41, Q. pou.] Justin L. 11.20, Luke 23.34 (v.l.), Q. 17.7, Matt. 28.19. allusive.