See [Endnote 204:1]

We may cut down the preliminary disquisitions as to the life and doctrines of Marcion, which have, indeed, a certain bearing upon the point at issue, but will be found given with sufficient fulness in 'Supernatural Religion,' or in any of the authorities. As in most other points relating to this period, there is some confusion in the chronological data, but these range within a comparatively limited area. The most important evidence is that of Justin, who, writing as a contemporary (about 147 A.D.) [Endnote 205:1], says that at that time Marcion had 'in every nation of men caused many to blaspheme' [Endnote 205:2]; and again speaks of the wide spread of his doctrines ([Greek: ho polloi peisthentes, k.t.l.]) [Endnote 205:3]. Taking these statements along with others in Irenaeus, Tertullian, and Epiphanius, modern critics seem to be agreed that Marcion settled in Rome and began to teach his peculiar doctrines about 139-142 A.D. This is the date assigned in 'Supernatural Religion' [Endnote 205:4]. Volkmar gives 138 A.D. [Endnote 205:5] Tischendorf, on the apologetic side, would throw back the date as far as 130, but this depends upon the date assigned by him to Justin's 'Apology,' and conflicts too much with the other testimony.

It is also agreed that Marcion himself did actually use a certain Gospel that is attributed to him. The exact contents and character of that Gospel are not quite so clear, and its relation to the Synoptic Gospels, and especially to our third Synoptic, which bears the name of St. Luke, is the point that we have to determine.

The Church writers, Irenaeus, Tertullian, and Epiphanius, without exception, describe Marcion's Gospel as a mutilated or amputated version of St. Luke. They contrast his treatment of the evangelical tradition with that pursued by his fellow-Gnostic, Valentinus [Endnote 205:6]. Valentinus sought to prove his tenets by wresting the interpretation of the Apostolic writings; Marcion went more boldly to work, and, having first selected his Gospel, our third Synoptic, cut out the passages both in it and in ten Epistles of St. Paul, admitted by him to be genuine, which seemed to conflict with his own system. He is also said to have made additions, but these were in any case exceedingly slight.

The statement of the Church writers should hardly, perhaps, be put aside quite so summarily as is sometimes done. The life of Irenaeus overlapped that of Marcion considerably, and there seems to have been somewhat frequent communication between the Church at Lyons, where he was first presbyter and afterwards bishop, and that of Rome, where Marcion was settled; but Irenaeus [Endnote 206:1], as well as Tertullian and Epiphanius, alludes to the mutilation of St. Luke's Gospel by Marcion as a notorious fact. Too much stress, however, must not be laid upon this, because the Catholic writers were certainly apt to assume that their own view was the only one tenable.

The modern controversy is more important, though it has to go back to the ancient for its data. The question in debate may be stated thus. Did Marcion, as the Church writers say, really mutilate our so-called St. Luke (the name is not of importance, but we may use it as standing for our third Synoptic in its present shape)? Or, is it not possible that the converse may be true, and that Marcion's Gospel was the original and ours an interpolated version? The importance of this may, indeed, be exaggerated, because Marcion's Gospel is at any rate evidence for the existence at his date in a collected form of so much of the third Gospel (rather more than two-thirds) as he received. Still the issue is not inconsiderable: for, upon the second hypothesis, if the editor of our present Gospel made use of that which was in the possession of Marcion, his date may be--though it does not follow that it certainly would be--thrown into the middle of the second century, or even beyond, if the other external evidence would permit; whereas, upon the first hypothesis, the Synoptic Gospel would be proved to be current as early as 140 A.D.; and there will be room for considerations which may tend to date it much earlier. There will still be the third possibility that Marcion's Gospel may be altogether independent of our present Synoptic, and that it may represent a parallel recension of the evangelical tradition. This would leave the date of the canonical Gospel undetermined.

It is a fact worth noting that the controversy, at least in its later and more important stages, had been fought, and, to all appearance, fought out, within the Tübingen school itself. Olshausen and Hahn, the two orthodox critics who were most prominently engaged in it, after a time retired and left the field entirely to the Tübingen writers.

The earlier critics who impugned the traditional view appear to have leaned rather to the theory that Marcion's Gospel and the canonical Luke are, more or less, independent offshoots from the common ground-stock of the evangelical narratives. Ritschl, and after him Baur and Schwegler, adopted more decidedly the view that the canonical Gospel was constructed out of Marcion's by interpolations directed against that heretic's teaching. The reaction came from a quarter whence it would not quite naturally have been expected--from one whose name we have already seen associated with some daring theories, Volkmar, Professor of Theology at Zürich. With him was allied the more sober-minded, laborious investigator, Hilgenfeld. Both these writers returned to the charge once and again. Volkmar's original paper was supplemented by an elaborate volume in 1852, and Hilgenfeld, in like manner, has reasserted his conclusions. Baur and Ritschl professed themselves convinced by the arguments brought forward, and retracted or greatly modified their views. So far as I am aware, Schwegler is the only writer whose opinion still stands as it was at first expressed; but for some years before his death, which occurred in 1857, he had left the theological field.

Without at all prejudging the question on this score, it is difficult not to feel a certain presumption in favour of a conclusion which has been reached after such elaborate argument, especially where, as here, there could be no suspicion of a merely apologetic tendency on either side. Are we, then, to think that our English critic has shown cause for reopening the discussion? There is room to doubt whether he would quite maintain as much as this himself. He has gone over the old ground, and reproduced the old arguments; but these arguments already lay before Hilgenfeld and Volkmar in their elaborate researches, and simply as a matter of scale the chapter in 'Supernatural Religion' can hardly profess to compete with these.

Supposing, for the moment, that the author has proved the points that he sets himself to prove, to what will this amount? He will have shown (a) that the patristic statement that Marcion mutilated St. Luke is not to be accepted at once without further question; (b) that we cannot depend with perfect accuracy upon the details of his Gospel, as reconstructed from the statements of Tertullian and Epiphanius; (c) that it is difficult to explain the whole of Marcion's alleged omissions, on purely, dogmatic grounds--assuming the consistency of his method.

With the exception of the first, I do not think these points are proved to any important extent; but, even if they were, it would still, I believe, be possible to show that Marcion's Gospel was based upon our third Synoptic by arguments which hardly cross or touch them at all.

But, before we proceed further, it is well that we should have some idea as to the contents of the Marcionitic Gospel. And here we are brought into collision with the second of the propositions just enunciated. Are we able to reconstruct that Gospel from the materials available to us with any tolerable or sufficient approach to accuracy? I believe no one who has gone into the question carefully would deny that we can. Here it is necessary to define and guard our statements, so that they may cover exactly as much ground as they ought and no more.

Our author quotes largely, especially from Volkmar, to show that the evidence of Tertullian and Epiphanius is not to be relied upon. When we refer to the chapter in which Volkmar deals with this subject [Endnote 209:1]--a chapter which is an admirable specimen of the closeness and thoroughness of German research--we do indeed find some such expressions, but to quote them alone would give an entirely erroneous impression of the conclusion to which the writer comes. He does not say that the statements of Tertullian and Epiphanius are untrustworthy, simply and absolutely, but only that they need to be applied with caution _on certain points_. Such a point is especially the silence of these writers as proving, or being supposed to prove, the absence of the corresponding passage in Marcion's Gospel. It is argued, very justly, that such an inference is sometimes precarious. Again, in quoting longer passages, Epiphanius is in the habit of abridging or putting an &c. ([Greek: kai ta hexaes-- kai ta loipa]), instead of quoting the whole. This does not give a complete guarantee for the intermediate portions, and leaves some uncertainty as to where the passage ends. Generally it is true that the object of the Fathers is not critical but dogmatic, to refute Marcion's system out of his own Gospel. But when all deductions have been made on these grounds, there are still ample materials for reconstructing that Gospel with such an amount of accuracy at least as can leave no doubt as to its character. The wonder is that we are able to do so, and that the statements of the Fathers should stand the test so well as they do. Epiphanius especially often shows the most painstaking care and minuteness of detail. He has reproduced the manuscript of Marcion's Gospel that he had before him, even to its clerical errors [Endnote 210:1]. He and Tertullian are writing quite independently, and yet they confirm each other in a remarkable manner. 'If we compare the two witnesses,' says Volkmar, 'we find the most satisfactory (sicher- stellendste) coincidence in their statements, entirely independent as they are, as well in regard to that which Marcion has in common with Luke, as in regard to very many of the points in which his text differed from the canonical. And this applies not only to simple omissions which Epiphanius expressly notes and Tertullian confirms by passing over what would otherwise have told against Marcion, but also to the minor variations of the text which Tertullian either happens to name or indicate by his translation, while they are confirmed by the direct statement of [the other] opponent who is equally bent on finding such differences' [Endnote 211:1]. Out of all the points on which they can be compared, there is a real divergence only in two. Of these, one Volkmar attributes to an oversight on the part of Epiphanius, and the other to a clerical omission in his manuscript [Endnote 211:2]. When we consider the cumbrousness of ancient MSS., the absence of divisions in the text, and the consequent difficulty of making exact references, this must needs be taken for a remarkable result. And the very fact that we have two--or, including Irenaeus, even three--independent authorities, makes the text of Marcion's Gospel, so far as those authorities are available, or, in other words, for the greater part of it, instead of being uncertain among quite the most certain of all the achievements of modern criticism [Endnote 211:3].

This is seen practically--to apply a simple test--in the large amount of agreement between critics of the most various schools as to the real contents of the Gospel. Our author indeed speaks much of the 'disagreement.' But by what standard does he judge? Or, has he ever estimated its extent? Putting aside merely verbal differences, the total number of whole verses affected will be represented in the following table:--

iv. 16-30: doubt as to exact extent of omissions affecting about half the verses.

38, 39: omitted according to Hahn; retained according to Hilgenfeld and Volkmar.

vii. 29-35: omitted, Hahn and Ritschl; retained, Hilgenfeld and Volkmar.

x. 12-15: ditto ditto.

xiii. 6-10: omitted, Volkmar; retained, Hilgenfeld and Rettig.

xvii. 5-10: omitted, Ritschl; retained, Volkmar and Hilgenfeld.

14-19: doubt as to exact omissions.

xix. 47, 48: omitted, Hilgenfeld and Volkmar; retained, Hahn and Anger.

xxii. 17, 18: doubtful.

23-27: omitted, Ritschl; retained, Hilgenfeld and Volkmar.

43, 44: ditto ditto.

xxiii. 39-42: ditto ditto.

47-49: omitted, Hahn; retained, Ritschl, Hilgenfeld, Volkmar.

xxiv. 47-53: uncertain [Endnote 212:1].

This would give, as a maximum estimate of variation, some 55 verses out of about 804, or, in other words, about seven per cent. But such an estimate would be in fact much too high, as there can be no doubt that the earlier researches of Hahn and Ritschl ought to be corrected by those of Hilgenfeld and Volkmar; and the difference between these two critics is quite insignificant. Taking the severest view that it is possible to take, no one will maintain that the differences between the critics are such as to affect the main issue, so that upon one hypothesis one theory would hold good, and upon another hypothesis another. It is a mere question of detail.

We may, then, reconstruct the Gospel used by Marcion with very considerable confidence that we have its real contents before us. In order to avoid any suspicion I will take the outline given in 'Supernatural Religion' (ii. p. 127), adding only the passage St. Luke vii. 29-35, which, according to the author's statement (a mistaken one, however) [Endnote 213:1], is 'generally agreed' to have been wanting in Marcion's Gospel. In that Gospel, then, the following portions of our present St. Luke were omitted:--

Chaps. i. and ii, including the prologue, the Nativity, and the birth of John the Baptist.

Chap. iii (with the exception of ver. 1), containing the baptism of our Lord, the preaching of St. John, and the genealogy.

iv. 1-13, 17-20, 24: the Temptation, the reading from Isaiah.

vii. 29-35: the gluttonous man.

xi. 29-32, 49-51: the sign of Jonas, and the blood of the prophets.

xiii. 1-9, 29-35: the slain Galileans, the fig-tree, Herod, Jerusalem.

xv. 11-32: the prodigal son.

xvii. 5-10: the servant at meat.

xviii. 31-34: announcement of the Passion.

xix. 29-48: the Triumphal Entry, woes of Jerusalem, cleansing of the Temple.

xx. 9-18, 37, 38: the wicked husbandmen; the God of Abraham.

xxi. 1-4, 18, 21, 22: the widow's mite; 'a hair of your head;' flight of the Church.

xxii. 16-18, 28-30, 35-38, 49-51: the fruit of the vine, 'eat at my table,' 'buy a sword,' the high-priest's servant.

xxiv. 47-53: the last commission, the Ascension.

Here we have another remarkable phenomenon. The Gospel stands to our Synoptic entirely in the relation of defect. We may say entirely, for the additions are so insignificant--some thirty words in all, and those for the most part supported by other authority--that for practical purposes they need not be reckoned. With the exception of these thirty words inserted, and some, also slight, alterations of phrase, Marcion's Gospel presents simply an abridgment of our St. Luke.

Does not this almost at once exclude the idea that they can be independent works? If it does not, then let us compare the two in detail. There is some disturbance and re-arrangement in the first chapter of Marcion's Gospel, though the substance is that of the third Synoptic; but from this point onwards the two move step by step together but for the omissions and a single transposition (iv. 27 to xvii. 18). Out of fifty-three sections peculiar to St. Luke--from iv. 16 onwards--all but eight were found also in Marcion's Gospel. They are found, too, in precisely the same order. Curious and intricate as is the mosaic work of the third Gospel, all the intricacies of its pattern are reproduced in the Gospel of Marcion. Where Luke makes an insertion in the groundstock of the narrative, there Marcion makes an insertion also; where Luke omits part of the narrative, Marcion does the same. Among the documents peculiar to St. Luke are some of a very marked and individual character, which seem to have come from some private source of information. Such, for instance, would be the document viii. 1-3, which introduces names so entirely unknown to the rest of the evangelical tradition as Joanna and Susanna [Endnote 215:1]. A trace of the same, or an allied document, appears in chap. xxiv, where we have again the name Joanna, and afterwards that of the obscure disciple Cleopas. Again, the mention of Martha and Mary is common only to St. Luke and the fourth Gospel. Zacchaeus is peculiar to St. Luke. Yet, not only does each of the sections relating to these personages re-appear in Marcion's Gospel, but it re-appears precisely at the same place. A marked peculiarity in St. Luke's Gospel is the 'great intercalation' of discourses, ix. 51 to xviii. 14, evidently inserted without regard to chronological order. Yet this peculiarity, too, is faithfully reproduced in the Gospel of Marcion with the same disregard of chronology--the only change being the omission of about forty-one verses from a total of three hundred and eighty. When Luke has the other two Synoptics against him, as in the insertions Matt. xiv. 3-12, Mark vi. 17-29, and again Matt. xx. 20-28, Mark x. 35-45, and Matt. xxi. 20-22, Mark xi. 20-26, Marcion has them against him too. Where the third Synoptist breaks off from his companions (Luke ix. 17, 18) and leaves a gap, Marcion leaves one too. It has been noticed as characteristic of St. Luke that, where he has recorded a similar incident before, he omits what might seem to be a repetition of it: this characteristic is exactly reflected in Marcion, and that in regard to the very same incidents. Then, wherever the patristic statements give us the opportunity of comparing Marcion's text with the Synoptic--and this they do very largely indeed--the two are found to coincide with no greater variation than would be found between any two not directly related manuscripts of the same text. It would be easy to multiply these points, and to carry them to any degree of detail; if more precise and particular evidence is needed it shall be forthcoming, but in the meantime I think it may be asserted with confidence that two alternatives only are possible. Either Marcion's Gospel is an abridgment of our present St. Luke, or else our present St. Luke is an expansion by interpolation of Marcion's Gospel, or of a document co-extensive with it. No third hypothesis is tenable.

It remains, then, to enquire which of these two Gospels had the priority--Marcion's or Luke's; which is to stand first, both in order of time and of authenticity. This, too, is a point that there are ample data for determining.

(1.) And, first, let us consider what presumption is raised by any other part of Marcion's procedure. Is it likely that he would have cut down a document previously existing? or, have we reason for thinking that he would be scrupulous in keeping such a document intact?

The author of 'Supernatural Religion' himself makes use of this very argument; but I cannot help suspecting that his application of it has slipped in through an oversight or misapprehension. When first I came across the argument as employed by him, I was struck by it at once as important if only it was sound. But, upon examination, not only does it vanish into thin air as an argument in support of the thesis he is maintaining, but there remains in its place a positive argument that tells directly and strongly against that thesis. A passage is quoted from Canon Westcott, in which it is stated that while Tertullian and Epiphanius accuse Marcion of altering the text of the books which he received, so far as his treatment of the Epistles is concerned this is not borne out by the facts, out of seven readings noticed by Epiphanius two only being unsupported by other authority. It is argued from this that Marcion 'equally preserved without alteration the text which he found in his manuscript of the Gospel.' 'We have no reason to believe the accusation of the Fathers in regard to the Gospel--which we cannot fully test-- better founded than that in regard to the Epistles, which we can test, and find unfounded' [Endnote 217:1]. No doubt the premisses of this argument are true, and so also is the conclusion, strictly as it stands. It is true that the Fathers accuse Marcion of tampering with the text in various places, both in the Epistles and in the Gospels where the allegation can be tested, and where it is found that the supposed perversion is simply a difference of reading, proved to be such by its presence in other authorities [Endnote 217:2]. But what is this to the point? It is not contended that Marcion altered to any considerable extent (though he did slightly even in the Epistles [Endnote 217:3]) the text _which he retained_, but that he mutilated and cut out whole passages from that text. He can be proved to have done this in regard to the Epistles, and therefore it is fair to infer that he dealt in the same way with the Gospel. This is the amended form in which the argument ought to stand. It is certain that Marcion made a large excision before Rom. xi. 33, and another after Rom. viii. 11; he also cut out the 'mentiones Abrahae' from Gal. iii. 7, 14, 16-18 [Endnote 218:1]. I say nothing about his excision of the last two chapters of the Epistle to the Romans, because on that point a controversy might be raised. But the genuineness of these other passages is undisputed and indisputable. It cannot be argued here that our text of the Epistle has suffered from later interpolation, and therefore, I repeat, it is so much the more probable that Marcion took from the text of the Gospel than that a later editor added to it.

(2.) In examining the internal evidence from the nature and structure of Marcion's Gospel, it has hitherto been the custom to lay most stress upon its dogmatic character. The controversy in Germany has turned chiefly on this. The critics have set themselves to show that the variations in Marcion's Gospel either could or could not be explained as omissions dictated by the exigencies of his dogmatic system. This was a task which suited well the subtlety and inventiveness of the German mind, and it has been handled with all the usual minuteness and elaboration. The result has been that not only have Volkmar and Hilgenfeld proved their point to their own satisfaction, but they also convinced Ritschl and partially Baur; and generally we may say that in Germany it seems to be agreed at the present time that the hypothesis of a mutilated Luke suits the dogmatic argument better than that of later Judaising interpolations.

I have no wish to disparage the results of these labours, which are carried out with the splendid thoroughness that one so much admires. Looking at the subject as impartially as I can, I am inclined to think that the case is made out in the main. The single instance of the perverted sense assigned to [Greek: kataelthen] in iv. 31 must needs go a long way. Marcion evidently intends the word to be taken in a transcendental sense of the emanation and descent to earth of the Aeon Christus [Endnote 219:1]. It is impossible to think that this sense is more original than the plain historical use of the word by St. Luke, or to mistake the dogmatic motive in the heretical recension. There is also an evident reason for the omission of the first chapters which relate the human birth of Christ, which Marcion denied, and one somewhat less evident, though highly probable, for the omission of the account of the Baptist's ministry, John being regarded as the finisher of the Old Testament dispensation--the work of the Demiurge. This omission is not quite consistently carried out, as the passage vii. 24-28 is retained--probably because ver. 28 itself seemed to contain a sufficient qualification. The genealogy, as well as viii. 19, was naturally omitted for the same reason as the Nativity. The narrative of the Baptism Marcion could not admit, because it supplied the foundation for that very Ebionism to which his own system was diametrically opposed. The Temptation, x. 21 ('Lord ... of earth'), xxii. 18 ('the fruit of the vine'), xxii. 30 ('eat and drink at my table'), and the Ascension, may have been omitted because they contained matter that seemed too anthropomorphic or derogatory to the Divine Nature. On the other hand, xi. 29-32 (Jonah and Solomon), xi. 49-51 (prophets and apostles), xiii. 1 sqq. (the fig-tree, as the Jewish people?), xiii. 31-35 (the prophet in Jerusalem), the prodigal son (perhaps?), the wicked husbandmen (more probably), the triumphal entry (as the fulfilment of prophecy), the announcement of the Passion (also as such), xxi. 21, 22 (the same), and the frequent allusions to the Old Testament Scriptures, seem to have been expunged as recognising or belonging to the kingdom of the Demiurge [Endnote 220:1]. Again, the changes in xiii. 28, xvi. 17, xx. 35, are fully in accordance with Marcion's system [Endnote 220:2]. The reading which Marcion had in xi. 22 is expressly stated to have been common to the Gnostic heretics generally. In some of these instances the dogmatic motive is gross and palpable, in most it seems to have been made out, but some (such as especially xiii. 1-9) are still doubtful, and the method of excision does not appear to have been carried out with complete consistency.

This, indeed, was only to be expected. We are constantly reminded that Tertullian, a man, with all his faults, of enormous literary and general power, did not possess the critical faculty, and no more was that faculty likely to be found in Marcion. It is an anachronism to suppose that he would sit down to his work with that regularity of method and with that subtle appreciation of the affinities of dogma which characterise the modern critic. The Septuagint translators betray an evident desire to soften down the anthropomorphism of the Hebrew; but how easy would it be to convict them of inconsistency, and to show that they left standing expressions as strong as any that they changed! If we judge Marcion's procedure by a standard suited to the age in which he lived, our wonder will be, not that he has shown so little, but so much, consistency and insight.

I think, therefore, that the dogmatic argument, so far as it goes, tells distinctly in favour of the 'mutilation' hypothesis. But at the same time it should not be pressed too far. I should be tempted to say that the almost exclusive and certainly excessive use of arguments derived from the history of dogma was the prime fallacy which lies at the root of the Tübingen criticism. How can it be thought that an Englishman, or a German, trained under and surrounded by the circumstances of the nineteenth century, should be able to thread all the mazes in the mind of a Gnostic or an Ebionite in the second? It is difficult enough for us to lay down a law for the actions of our own immediate neighbours and friends; how much more difficult to 'cast the shell of habit,' and place ourselves at the point of view of a civilisation and world of thought wholly different from our own, so as not only to explain its apparent aberrations, but to be able to say, positively, 'this must have been so,' 'that must have been otherwise.' Yet such is the strange and extravagant supposition that we are assumed to make. No doubt the argument from dogma has its place in criticism; but, on the whole, the literary argument is safer, more removed from the influence of subjective impressions, more capable of being cast into a really scientific form.

(3.) I pass over other literary arguments which hardly admit of this form of expression--such as the improbability that the Preface or Prologue was not part of the original Gospel, but a later accretion; or, again, from Marcion's treatment of the Synoptic matter in the third Gospel, both points which might be otherwise worth dilating upon. I pass over these, and come at once, without further delay, to the one point which seems to me really to decide the character of Marcion's Gospel and its relation to the Synoptic. The argument to which I allude is that from style and diction. True the English mind is apt to receive literary arguments of that kind with suspicion, and very justly so long as they rest upon a mere vague subjective _ipse dixit_; but here the question can be reduced to one of definite figures and of weighing and measuring. Bruder's Concordance is a dismal- looking volume--a mere index of words, and nothing more. But it has an eloquence of its own for the scientific investigator. It is strange how clearly many points stand out when this test comes to be applied, which before had been vague and obscure. This is especially the case in regard to the Synoptic Gospels; for, in the first place, the vocabulary of the writers is very limited and similar phrases have constant tendency to recur, and, in the second place, the critic has the immense advantage of being enabled to compare their treatment of the same common matter, so that he can readily ascertain what are the characteristic modifications introduced by each. Dr. Holtzmann, following Zeller and Lekebusch, has made a full and careful analysis of the style and vocabulary of St. Luke [Endnote 223:1], but of course without reference to the particular omissions of Marcion. Let us then, with the help of Bruder, apply Holtzmann's results to these omissions, with a view to see whether there is evidence that they are by the same hand as the rest of the Gospel.

It would be beyond the proportions of the present enquiry to exhibit all the evidence in full. I shall, therefore, not transcribe the whole of my notes, but merely give a few samples of the sort of evidence producible, along with a brief summary of the general results.

Taking first certain points by which the style of the third Evangelist is distinguished from that of the first in their treatment of common matter, Dr. Holtzmann observes, that where Matthew has [Greek: grammateus], Luke has in six places the word [Greek: nomikos], which is only found three times besides in the New Testament (once in St. Mark, and twice in the Epistle to Titus). Of the places where it is used by St. Luke, one is the omitted passage, vii. 30. In citations where Matthew has [Greek: to rhaethen] (14 times; not at all in Luke), Luke prefers the perfect form [Greek: to eiraemenon], so in ii. 24 (Acts twice); compare [Greek: eiraetai], iv. 21. Where Matthew has [Greek: arti] (7 times), Luke has always [Greek: nun], never [Greek: arti]: [Greek: nun] is used in the following passages, omitted by Marcion: i. 48, ii. 29, xix. 42, xxii. 18, 36. With Matthew the word [Greek: eleos] is masculine, with Luke neuter, so five times in ch. i. and in x. 37, which was retained by Marcion.

Among the peculiarities of style noted by Dr. Holtzmann which recur in the omitted portions the following are perhaps some of the more striking. Peculiar use of [Greek: to] covering a whole phrase, i. 62 [Greek: to ti an theloi kaleisthai], xix. 48, xxii. 37, and five other places. Peculiar attraction of the relative with preceding case of [Greek: pas], iii. 19, xix. 37, and elsewhere. The formula [Greek: elege (eipe) de parabolaen] (not found in the other Synoptics), xiii. 6, xx. 9, 19, and ten times besides. [Greek: Tou] pleonastic with the infinitive, once in Mark, six times in Matthew, twenty-five times in Luke, of which three times in chap. i, twice in chap. ii, iv. 10, xxi. 22. Peculiar combinations with [Greek: kata, kata to ethos, eiothos, eithismenon], i. 9, ii. 27, 42, and twice. [Greek: Kath' haemeran], once in the other Gospels, thirteen times in Luke and Acts xix. 47; [Greek: kat' etos], ii. 41; [Greek: kata] with peculiar genitive of place, iv. 14 (xxiii. 5) [Endnote 224:1]. Protasis introduced by [Greek: kai hote], ii. 21, 22, 42, [Greek: kai hos], ii. 39, xv. 25, xix. 41. Uses of [Greek: egeneto], especially with [Greek: en to] and infinitive, twice in Mark, in Luke twenty-two times, i. 8, ii. 6, iii. 21, xxiv. 51; [Greek: en to] with the infinitive, three times in St. Matthew, once in St. Mark, thirty-seven times in St. Luke, including i. 8, 21, ii. 6, 27, 43, iii. 21. Adverbs: [Greek: exaes] and [Greek: kathexaes], ten times in the third Gospel and the Acts alone in the New Testament, i. 3; [Greek: achri], twenty times in the third Gospel and Acts, only once in the other Gospels, i. 20, iv. 13; [Greek: exaiphnaes], four times in the Gospel and Acts, once besides in the New Testament, ii. 13; [Greek: parachraema], seventeen times in the Gospel and Acts, twice in the rest of the New Testament, i. 64; [Greek: en meso], thirteen times in the Gospel and Acts, five times in the other Synoptics, ii. 46, xxi. 21. Fondness for optative in indirect constructions, i. 29, 62, iii. 15, xv. 26. Peculiar combination of participles, ii. 36 ([Greek: probebaekuia zaesasa]), iii. 23 ([Greek: archomenos on]), iv. 20 ([Greek: ptuxas apodous]), very frequent. [Greek: Einai], with participle for finite verb (forty-eight times in all), i. 7, 10, 20, 21, 22, ii. 8, 26, 33, 51, iii. 23, iv. 16 ([Greek: aen tethrammenos], omitted by Marcion), iv. 17, 20, xv. 24, 32, xviii. 34, xix. 47, xx. 17, xxiv. 53. Construction of [Greek: pros] with accusative after [Greek: eipein, lalein, apokrinesthai], frequent in Luke, rare in the rest of the New Testament, i. 13, 18, 19, 28, 34, 55, 61, 73, ii. 15, 18, 34, 48, 49, iii. 12, 13, 14, iv. 4, xiii. 7, 34, xv. 22, xviii. 31, xix. 33, 39, xx. 9, 14, 19. This is thrown into marked relief by the contrast with the other Synoptics; the only two places where Matthew appears to have the construction are both ambiguous, iii. 15 (doubtful reading, probably [Greek: auto]), and xxvii. 14 ([Greek: apekrithae auto pros oude hen rhaema]). No other evangelist speaks so much of [Greek: Pneuma hagion], i. 15, 35, 41, 67, ii. 25, 66, iii. 16, 22, iv. 1 (found also in Marcion's reading of xi. 2). Peculiar use of pronouns: Luke has the combination [Greek: kai autos] twenty-eight times, Matthew only twice (one false reading), Mark four or perhaps five times, i. 17, 22, ii. 28, iii. 23, xv. 14; [Greek: kai autoi] Mark has not at all, Matthew twice, Luke thirteen times, including ii. 50, xviii. 34, xxiv. 52.

We now come to the test supplied by the vocabulary. The following are some of the words peculiar to St. Luke, or found in his writings with marked and characteristic frequency, which occur in those parts of our present Gospel that were wanting in Marcion's recension: [Greek: anestaen, anastas] occur three times in St. Matthew, twice in St. John, four times in the writings of St. Paul, twenty-six times in the third Gospel and thirty-five times in the Acts, and are found in i. 39, xv. 18, 20; [Greek: antilegein] appears in ii. 34, five times in the rest of the Gospel and the Acts, and only four times together in the rest of the New Testament; [Greek: hapas] occurs twenty times in the Gospel, sixteen times in the Acts, only ten times in the rest of the New Testament, but in ii. 39, iii. 16, 21, iv. 6, xv. 13, xix. 37, 48, xxi. 4 (bis); three of these are, however, doubtful readings. [Greek: aphesis ton amartion], ten times in the Gospel and Acts, seven times in the rest of the New Testament, i. 77, iii. 3. [Greek: dei], Dr. Holtzmann says, 'is found more often in St. Luke than in all the other writers of the New Testament put together.' This does not appear to be strictly true; it is, however, found nineteen times in the Gospel and twenty-five times in the Acts to twenty-four times in the three other Gospels; it occurs in ii. 49, xiii. 33, xv. 32, xxii. 37. [Greek: dechesthai], twenty-four times in the Gospel and Acts, twenty-six times in the rest of the New Testament, six times in St. Matthew, three in St. Mark, ii. 28, xxii. 17. [Greek: diatassein], nine times in the Gospel and Acts, seven times in the rest of the New Testament (Matthew once), iii. 13, xvii. 9, 10. [Greek: dierchesthai] occurs thirty-two times in the Gospel and Acts, twice in each of the other Synoptics, and eight times in the rest of the New Testament, and is found in ii. 15, 35. [Greek: dioti], i. 13, ii. 7 (xxi. 28, and Acts, not besides in the Gospels). [Greek: ean], xxii. 51 (once besides in the Gospel, eight times in the Acts, and three times in the rest of the New Testament). [Greek: ethos], i. 9, ii. 42, eight times besides in St. Luke's writings and only twice in the rest of the New Testament. [Greek: enantion], five times in St. Luke's writings, once besides, i. 8. [Greek: enopion], correcting the readings, twenty times in the Gospel, fourteen times in the Acts, not at all in the other Synoptists, once in St. John, four times in chap. i, iv. 7, xv. 18, 21 (this will be noticed as a very remarkable instance of the extent to which the diction of the third Evangelist impressed itself upon his writings). [Greek: epibibazein], xix. 35 (and twice, only by St. Luke). [Greek: epipiptein], i. 12, xv. 20 (eight times in the Acts and three times in the rest of the New Testament). [Greek: ai eraemoi], only in St. Luke, i. 80, and twice. [Greek: etos] (fifteen times in the Gospel, eleven times in the Acts, three times in the other Synoptics and three times in St. John), four times in chap. ii, iii. 1, 23, xiii, 7, 8, xv. 29. [Greek: thaumazein epi tini], Gospel and Acts five times (only besides in Mark xii. 17), ii. 33. [Greek: ikanos] in the sense of 'much,' 'many,' seven times in the Gospel, eighteen times in the Acts, and only three times besides in the New Testament, iii. 16, xx. 9 (compare xxii. 38). [Greek: kathoti] (like [Greek: kathexaes] above), is only found in St. Luke's writings, i. 7, and five times in the rest of the Gospel and the Acts. [Greek: latreuein], 'in Luke, much oftener than in other parts of the New Testament,' i. 74, ii. 37, iv. 8, and five times in the Acts. [Greek: limos], six times in the Gospel and Acts, six times in the rest of the New Testament, xv. 14, 17. [Greek: maen] (month), i. 24, 26, 36, 56 (iv. 25), alone in the Gospels, in the Acts five times. [Greek: oikos] for 'family,' i. 27, 33, 69, ii. 4, and three times besides in the Gospel, nine times in the Acts. [Greek: plaethos] (especially in the form [Greek: pan to plaethos]), twenty-five times in St. Luke's writings, seven times in the rest of the New Testament, 1. 19, ii. 13, xix. 37. [Greek: plaesai, plaesthaenai], twenty-two times in St. Luke's writings, only three times besides in the New Testament, i. 15, 23, 41, 57, 67, ii. 6, 21, 22, xxi. 22. [Greek: prosdokan], eleven times in the Gospel and Acts, five times in the rest of the New Testament (Matthew twice and 2 Peter), i. 21, iii. 15. [Greek: skaptein], only in Luke three times, xiii. 8. [Greek: speudein], except in 2 Peter iii. 12, only in St. Luke's writings, ii. 16. [Greek: sullambanein], ten times in the Gospel and Acts, five times in the rest of the New Testament, i. 24, 31, 36, ii. 21. [Greek: sumballein], only in Lucan writings, six times, ii. 19. [Greek: sunechein], nine times in the Gospel and Acts, three times besides in the New Testament, xix. 43. [Greek: sotaeria], in chap. i. three times, in the rest of the Gospel and Acts seven times, not in the other Synoptic Gospels. [Greek: hupostrephein], twenty-two times in the Gospel, eleven times in the Acts, and only five times in the rest of the New Testament (three of which are doubtful readings), i. 56, ii. 20, 39, 43, 45, iv. 1, (14), xxiv. 52. [Greek: hupsistos] occurs nine times in the Gospel and Acts, four times in the rest of the New Testament, i. 32, 35, 76, ii. 14, xix. 38. [Greek: hupsos] is also found in i. 78, xxiv. 49. [Greek: charis] is found, among the Synoptics, only in St. Luke, eight times in the Gospel, seventeen times in the Acts, i. 30, ii. 40, 52, xvii. 9. [Greek: hosei] occurs nineteen times in the Gospel and Acts (four doubtful readings, of which two are probably false), seventeen times in the rest of the New Testament (ten doubtful readings, of which in the Synoptic Gospels three are probably false), i. 56, iii. 23.

It should be remembered that the above are only samples from the whole body of evidence, which would take up a much larger space if exhibited in full. The total result may be summarised thus. Accepting the scheme of Marcion's Gospel given some pages back, which is substantially that of 'Supernatural Religion,' Marcion will have omitted a total of 309 verses. In those verses there are found 111 distinct peculiarities of St. Luke's style, numbering in all 185 separate instances; there are also found 138 words peculiar to or specially characteristic of the third Evangelist, with 224 instances. In other words, the verified peculiarities of St. Luke's style and diction (and how marked many of these are will have been seen from the examples above) are found in the portions of the Gospel omitted by Marcion in a proportion averaging considerably more than one to each verse! [Endnote 229:1] Coming to detail, we find that in the principal omission-- that of the first two chapters, containing 132 verses--there are 47 distinct peculiarities of style, with 105 instances; and 82 characteristic words, with 144 instances. In the 23 verses of chap. iii. omitted by Marcion (for the genealogy need not be reckoned), the instances are 18 and 14, making a total of 32. In 18 verses omitted from chap. iv. the instances are 13 and 8 = 21. In another longer passage--the parable of the prodigal son--the instances are 8 of the first class and 20 of the second. In 20 verses omitted from chap. xix. the instances are 11 and 6; and in 11 verses omitted from chap. xx, 9 and 8. Of all the isolated fragments that Marcion had ejected from his Gospel, there are only four--iv. 24, xi. 49-51, xx. 37, 38, xxii. 28-30, nine verses in all--in which no peculiarities have been noticed. And yet even here the traces of authorship are not wanting. It happens strangely enough that in a list of parallel passages given by Dr. Holtzmann to illustrate the affinities of thought between St. Luke and St. Paul, two of these very passages--xi. 49 and xx. 38-- occur. I had intended to pursue the investigation through these resemblances, but it seems superfluous to carry it further.

It is difficult to see what appeal can be made against evidence such as this. A certain allowance should indeed be made for possible errors of computation, and some of the points may have been wrongly entered, though care has been taken to put down nothing that was not verified by its preponderating presence in the Lucan writings, and especially by its presence in that portion of the Gospel which Marcion undoubtedly received. But as a rule the method applies itself mechanically, and when every deduction has been made, there will still remain a mass of evidence that it does not seem too much to describe as overwhelming.

(4.) We may assume, then, that there is definite proof that the Gospel used by Marcion presupposes our present St. Luke, in its complete form, as it has been handed down to us. But when once this assumption has been made, another set of considerations comes in, which also carry with them an important inference. If Marcion's Gospel was an extract from a manuscript containing our present St. Luke, then not only is it certain that that Gospel was already in existence, but there is further evidence to show that it must have been in existence for some time. The argument in this case is drawn from another branch of Biblical science to which we have already had occasion to appeal--text-criticism. Marcion's Gospel, it is known, presents certain readings which differ both from the received and other texts. Some of these are thought by Volkmar and Hilgenfeld to be more original and to have a better right to stand in the text than those which are at present found there. These critics, however, base their opinion for the most part on internal grounds, and the readings defended by them are not as a rule those which are supported by other manuscript authority. It is to this second class rather that I refer as bearing upon the age of the canonical Gospel. The most important various readings of the existence of which we have proof in Marcion's Gospel are as follows [Endnote 231:1]:--

v. 14. The received (and best) text is [Greek: eis marturion autois]. Marcion, according to the express statement of Epiphanius (312 B), read [Greek: hina ae morturion touto humin], which is confirmed by Tertullian, who gives (_Marc._ iv. 8) 'Ut sit vobis in testimonium.' The same or a similar reading is found in D, [Greek: hina eis marturion ae humin touto], 'ut sit in testimonium vobis hoc,' d; 'ut sit in testimonium (--monia, ff) hoc vobis,' a (Codex Vercellensis), b (Codex Veronensis), c (Codex Colbertinus), ff (Codex Corbeiensis), l (Codex Rhedigerianus), of the Old Latin [Endnote 231:2].

v. 39 was _probably_ omitted by Marcion (this is inferred from the silence of Tertullian by Hilgenfeld, p. 403, and Rönsch, p. 634). The verse is also omitted in D, a, b, c, d, e, ff.

x. 22. Marcion's reading of this verse corresponded with that of other Gnostics, but has no extant manuscript authority. We have touched upon it elsewhere.

x. 25. [Greek: zoaen aionion], Marcion omitted [Greek: aionion] (Tert. Adv. Marc. iv. 25); so also the Old Latin Codex g'2 (San Germanensis).

xi. 2. Marcion read [Greek: eltheto to hagion pneuma sou eph' haemas] (or an equivalent; see Rönsch, p. 640) either for the clause [Greek: hagiasthaeto to onoma sou] or for [Greek: genaethaeto to thelaema sou], which is omitted in B, L, 1, Vulg., ff, Syr. Crt. There is a curious stray [Greek: eph' haemas] in D which may conceivably be a trace of Marcion's reading.

xii. 14. Marcion (and probably Tertullian) read [Greek: kritaen] (or [Greek: dikastaen]) only for [Greek: kritaen ae meristaen]; so D, a ('ut videtur,' Tregelles), c, Syr. Crt.

xii. 38. Marcion had [Greek: tae hesperinae phulakae] for [Greek: en tae deutera phulakae kai en tae tritae phulakae]. So b: D, c, e, ff, i, Iren. 334, Syr. Crt., combine the two readings in various ways.

xvi. 12. Marcion read [Greek: emon] for [Greek: humeteron]. So e (Palatinus), i (Vindobonensis), l (Rhedigerianus). [Greek: haemeteron] B. L, Origen.

xvii. 2. Marcion inserted the words [Greek: ouk egennaethae ae] (Tert. iv. 35), 'ne nasceretur aut,' a, b, c, ff, i, l.

xviii. 19. Here again Marcion had a variation which is unsupported by manuscript authority, but has to some extent a parallel in the Clementine Homilies, Justin, &c.

xxi. 18. was omitted by Marcion (Epiph. 316 B), and is also omitted in the Curetonian Syriac.

xxi. 27. Tertullian (iv. 39) gives the reading of Marcion as 'cum plurima virtute' = [Greek: meta dunameos pollaes [kai doxaes]], for [Greek: meta dun. k. dox. pollaes]; so D ([Greek: en dun. pol.]), and approximately Vulg., a, c, e, f, ff, Syr. Crt., Syr. Pst.

xxiii. 2. Marcion read [Greek: diastrephonta to ethnos kai katalionta ton nomon kai tous prophaetas kai keleuonta phorous mae dounai kai anastrephonta tas gunaikas kai ta tekna] (Epiph., 316 D), where [Greek: kataluonta ton nomon kai tous prophaetas] and [Greek: anastrephonta tas gunaikas kai ta tekna] are additions to the text, and [Greek: keleuonta phorous mae dounai] is a variation. Of the two additions the first finds support in b, (c), e, (ff), i, l; the second is inserted, with some variation, by c and e in verse 5.

We may thus tabulate the relation of Marcion to these various authorities. The brackets indicate that the agreement is only approximate. Marcion agrees with--

D, d, v. 14, v. 39; xii. 14, (xii. 28), (xxi. 27).

a (Verc.), v. 14, v. 39, xii. 14 (apparently), xvii. 2, (xxi. 27).

b (Ver.), v. 14, v. 39. xii. 38, xvii. 2, (xxiii. 2).

c (Colb.), v. 14, v. 39, xii. 14, (xii. 38), xvii. 2, (xxi. 27), (xxiii. 2), (xxiii. 2).

e (Pal.), v. 39, (xii. 38), xvi. 12, (xxi. 27), xxiii. 2, (xxiii. 2).

ff (Corb.), v. 14, v. 39, (xii. 38), xvii. 2, (xxi. 27), (xxiii. 2).

g'2 (Germ.), x. 25.

i (Vind.), (xii. 38), xvi. 12, xvii. 2, xxiii. 2.

l (Rhed.), v. 14, xvi. 12, xvii. 2, xiii. 2.

Syr. Crt., xii. 14, (xii. 38), xxi. 18, (xxi. 27).

It is worth noticing that xxii. 19 b, 20 (which is omitted in D, a, b, c, ff, i, l) appears to have been found in Marcion's Gospel, as in the Vulgate, c, and f (see Rönsch, p. 239). [Greek: apo tou mnaemeiou] in xxiv. 9 is also found (Rönsch, p. 246), though omitted by D, a, b, c, e, ff, l. There is no evidence to show whether the additions in ix. 55, xxiii. 34, and xxii. 43, 44 were present in Marcion's Gospel or not.

It will be observed that the readings given above have all what is called a 'Western' character. The Curetonian Syriac is well known to have Western affinities [Endnote 233:1]. Codd. a, b, c, and the fragment of i which extends from Luke x. 6 to xxiii. 10, represent the most primitive type of the Old Latin version; e, ff, and I give a more mixed text. As we should expect, the revised Latin text of Cod. f has no representation in Marcion's Gospel [Endnote 233:2].

These textual phenomena are highly interesting, but at the same time an exact analysis of them is difficult. No simple hypothesis will account for them. There can be no doubt that Marcion's readings are, in the technical sense, false; they are a deviation from the type of the pure and unadulterated text. At a certain point, evidently of the remotest antiquity, in the history of transcription, there was a branching off which gave rise to those varieties of reading which, though they are not confined to Western manuscripts, still, from their preponderance in these, are called by the general name of 'Western.' But when we come to consider the relations among those Western documents themselves, no regular descent or filiation seems traceable. Certain broad lines indeed we can mark off as between the earlier and later forms of the Old Latin, though even here the outline is in places confused; but at what point are we to insert that most remarkable document of antiquity, the Curetonian Syriac? For instance, there are cases (e.g. xvii. 2, xxiii. 2) where Marcion and the Old Latin are opposed to the Old Syriac, where the latter has undoubtedly preserved the correct reading. To judge from these alone, we should naturally conclude that the Syriac was simply an older and purer type than Marcion's Gospel and the Latin. But then again, on the other hand, there are cases (such as the omission of xxi. 18) where Marcion and the Syriac are combined, and the Old Latin adheres to the truer type. This will tend to show that, even at that early period, there must have been some comparison and correction--a _con_vergence as well as a _di_vergence-- of manuscripts, and not always a mere reproduction of the particular copy which the scribe had before him; at the same time it will also show that Marcion's Gospel, so far from being an original document, has behind it a deep historical background, and stands at the head of a series of copies which have already passed through a number of hands, and been exposed to a proportionate amount of corruption. Our author is inclined to lay stress upon the 'slow multiplication and dissemination of MSS.' Perhaps he may somewhat exaggerate this, as antiquarians give us a surprising account of the case and rapidity with which books were produced by the aid of slave-labour [Endnote 235:1]. But even at Rome the publishing trade upon this large scale was a novelty dating back no further than to Atticus, the friend of Cicero, and we should naturally expect that among the Christians--a poor and widely scattered body, whose tenets would cut them off from the use of such public machinery--the multiplication of MSS. would be slower and more attended with difficulty. But the slower it was the more certainly do such phenomena as these of Marcion's text throw back the origin of the prototype from which that text was derived. In the year 140 A.D. Marcion possesses a Gospel which is already in an advanced stage of transcription--which has not only undergone those changes which in some regions the text underwent before it was translated into Latin, but has undergone other changes besides. Some of its peculiarities are not those of the earliest form of the Latin version, but of that version in what may be called its second stage (e.g. xvi. 12). It has also affinities to another version kindred to the Latin and occupying a similar place to the Old Latin among the Churches of Syria. These circumstances together point to an antiquity fully as great as any that an orthodox critic would claim.

It should not be thought that because such indications are indirect they are therefore any the less certain. There is perhaps hardly a single uncanonical Christian document that is admittedly and indubitably older than Marcion; so that direct evidence there is naturally none. But neither is there any direct evidence for the antiquity of man or of the earth. The geologist judges by the fossils which he finds embedded in the strata as relics of an extinct age; so here, in the Gospel of Marcion, do we find relics which to the initiated eye carry with them their own story.

Nor, on the other hand, can it rightly be argued that because the history of these remains is not wholly to be recovered, therefore no inference from them is possible. In the earlier stages of a science like palaeontology it might have been argued in just the same way that the difficulties and confusion in the classification invalidated the science along with its one main inference altogether. Yet we can see that such an argument would have been mistaken. There will probably be some points in every science which will never be cleared up to the end of time. The affirmation of the antiquity of Marcion's Gospel rests upon the simple axiom that every event must have a cause, and that in order to produce complicated phenomena the interaction of complicated causes is necessary. Such an assumption involves time, and I think it is a safe proposition to assert that, in order to bring the text of Marcion's Gospel into the state in which we find it, there must have been a long previous history, and the manuscripts through which it was conveyed must have parted far from the parent stem.