Milligan on TC

W. Milligan

Modern Introduction

---

---

W. Milligan on TC (1873)

Excerpts: Part II

The State of Textual Criticism

"It is too generally imagined that we have no principles at all. Even intelligent persons are frequently under the impression that the science of textual criticism is a mere collection of undigested facts, that the textual critic has no fixed laws to regulate his procedure, that he works at random, and that the text eventually constructed by him is the result of arbitrary hypothesis or unregulated caprice. It will be our effort to show that this is not the case; and that, whatever uncertainty may still rest upon some parts of our subject, enough is known to make it strictly a science."

---

1. This optimistic fantasy was quickly refuted by S. W. Whitney. Even 20 years later, the state of NT Textual Criticism was obviously still quite a mess:

---

---

Weighing Sources

Differing Values (p.86-87)

It has been already stated that there are three sources from which these readings are to be derived, manuscripts, versions [early translations], and the writings of the early Fathers.

Had we reason to believe that all these authorities were of equal value, our course would be a simple one. Looking at them as so many witnesses, each entitled to the same degree of credit, we should simply reckon up the number upon opposing sides, and pronounce our verdict according to the numerical majority.

Such a state of things, however, is never exhibited in a court of justice. The value of evidence given by different witnesses differs very materially there.

The same principle comes into operation [here]. One of the first things that strikes the student of the NT text is the degree to which the many witnesses differ from one another [in value].

It is clear then that to merely [count] our witnesses will not do. We must distinguish their individual values. We must arrange and classify them. ...Were we to be guided by the number of witnesses [only] on either side, we would at once have to favour of the Received Text.

[But] examples may suffice to show how a decision in favour of a minority of witnesses may commend itself to the spiritually enlightened judgment. They will confirm the necessity of classification of our witnesses.

---

II: Grading Types of Evidence

Fathers, Versions, MSS (p.92-6)

We have first to make a comparative estimate of these three classes [Fathers, Versions, MSS].

1. Early Fathers: The autographs [of these] have long since perished. It cannot be doubted that, in their transcription, all the liabilities to error which affected the transcription of the inspired NT autographs would exist in even greater force. No such profound reverence was entertained for them as for Scripture.

Suppose a quotation given by a Father was different from familiar manuscripts: the Father had quoted it wrong, and justice both to the Bible and to him, required the error be corrected. Add to this that copies aren't so close to the time of composition, and these copies are few. The text then is more difficult to determine than with the NT.

But given a citation from an early Father, the question remains: Has it been adequately represented? Fathers were often loose quoting scripture. Whether careless, or finding passages hard to locate, or simply depending upon tradition, they were less careful of wording. They were often satisfied with the sense, rather than exact quotes. The same father quotes the same verse different ways.

Sometimes patristic citations are valuable, and given with intent to quote the original with accuracy. Arguments may be based on a single word, showing its existance and preference. There they can be valid tests for variants, confirming other witnesses.

But from all this, the highest rating possible is still secondary to MSS, and they are only subsidiary aids to establishing the true reading. They are not primary authorities.

2. Versions (early translations): These have great value through their testimony to the general purity and integrity of the form in which the NT has come down to us. This is due to providential early translation, preventing any extensive corruption, and demands admiration.

What is the relation of versions to manuscripts in fixing particular words or clauses of the Greek text? Just as in the case of citations, their position is subordinate.

The same problems confront us: What is the true text of the version itself? Early exposure to the same mischances make recovery as difficult. Add the temptation to correct from the Greek. Supposing the original wording, the host language may be widely different. The sense of the Greek may be mistaken.

On the other hand, some points have great weight: Translations can clearly testify to the presence or absence of whole clauses. Excessive literalness and habit assists identifying underlying Greek words and word-order. Isolated communities establish independant lines of transmission.

Translations can be a powerful aid in settling disputed texts; but they cannot be depended on as primary authorities either.

3. Manuscripts: We are forced back to manuscripts of the Greek text itself, upon documents directly descended from the ancient autographs. We may find other readings in patristic citations, or versions; but the instances must be rare indeed, for them to be regarded as legitimate without Greek manuscript support.

---

III: Second Stage - Weighing Greek MSS

Recensions, Ancient Readings, Text-types (p.96)

Greek manuscripts alone occupy the first rank. However, they may not all be of the same value.

It might seem then that the older they are, the more valuable they will be; and that, since the uncial manuscripts are our oldest group, we should depend mainly upon them.

However, there is a very great difference between the oldness of a manuscript and the oldness of its text. An original book belongs to the exact period at which its author lived, but not necessarily with copies. An 11th cent. cursive manuscript may present a lost text belonging to the 4th century, or even earlier. If so, it is clear that it is not to have its value diminished by the unimportant fact that it is written in cursive script, not uncial. No doubt an uncial is usually in fact old. But a cursive needs to be carefully evaluated. Outward appearance is inconclusive for the value of a MS.

One means of classifying MSS based on age of text is as follows:

(1) Identifying MSS belonging to Recensions:

First, we can separate our most ancient uncial manuscripts from the rest, and determine from them the general character of an ancient text. They tell us the state of the text in their day, and it must be old because they are old. We can then compare this text with more modern manuscripts.

If there is a very great resemblance, but the modern differences can only be explained as early Church readings, not simply edits over time, we judge the modern MSS ancient in text.

But if the differences are accounted by gradual process of change, typical of repeated copying, the MSS are modern in text as well as form.

Now it is often possible to trace ... the process of change going on, alterations, roughnesses softened down, difficulties cleared away, anomalous language removed, the whole character of the text becoming different from what it was. Wherever this can be done, we have gained the line of demarcation we are in search of. We are entitled to say that MSS with such changes are not so worthy of reliance as those which have escaped.

(2) Identifying MSS with Ancient Readings: (p.99)

In the second place, we can determine with still greater precision whether the text of any manuscript (MS) before us, be it uncial or cursive, is ancient or not.

We know from entirely independent testimony, how certain passages of Scripture were read in ancient times. By these we can test our witnesses. Say a MS has all the appearance of antiquity. We ask, How does it read such and such passages? If as they were read in ancient times; well. But if the readings are as we know that they were read only in modern times, we at once say, this is one of the later descendants.

Again, another MS comes before us with a modern appearance. We apply the same test. It may display a text that belongs to remote antiquity.

(3) Grouping MSS by text-type: (p.100)

In the third place, such a separation as that of which we have been speaking has actually been made in certain cases with the consent of all. Inquirers differ as to the value they attach to the classes thus separated from each other. They do not deny that the classes exist. There are modern manuscripts which all allow to possess an ancient text, and to be in this respect distinguished from the great mass of their fellows.

It is no doubt possible that even manuscripts, not exhibiting what may be called an ancient text upon the whole, may occasionally preserve an apostolic word or phrase that somehow or other has dropped out of their more ancient companions.

That they have preserved much that is apostolic is obvious, for in much all our witnesses agree, and that must be apostolic. They may therefore have preserved more. The difficulty is to make sure that they have done so.

Number or Quantity of MSS/Witnesses Discounted:

No agreement among themselves can be sufficient to convince us, for they may have been led astray by some common cause. No mere number witnessing the same thing can be of the least avail.

Unless we are satisfied that each individual of the number would be a good witness though he stood alone, the mere bringing them together as a multitude may disturb, but cannot convince. They must prove their title to be heard; and appearances being certainly against them, inasmuch as they differ from our older witnesses, we are entitled to ask that the proof they offer us shall be unambiguous and weighty.

Majority of Manuscripts Downgraded:

Thus then we have taken another step, and one most fruitful of results. It is a demonstrated fact, that the great mass of manuscripts belonging to the later centuries of the Christian Church cannot stand the tests of which we have been speaking. We shall not say that they are therefore to be put wholly aside, but certainly they are not primary authorities. Our circle has again been greatly narrowed.

---

IV: Third Step in Classification - Part A

State of Text, Rating Quality of MSS (p. 101 fwd)

In the process of classification so far we have been enabled to diminish greatly the number of our primary witnesses to the Greek text of the NT.

We have also established that all primary witnesses should have an ancient text, a text at least in general character. It may be thought that this should be enough; but it must be borne in mind that the oldest of our manuscripts does not go back to a point of time older than the first half of the 4th century, and that the texts therefore hitherto mainly used by us take us only to that date, 3 centuries later than the autographs.

This indeed would be a matter of no consequence if two conditions affecting the question before us could be fulfilled.

First, if, up to the period to which our oldest manuscripts belong, the text of the NT had remained pure and free from error, then those witnesses would at the same time give the text of the autographs.

Secondly, were our witnesses agreed in their testimony, we should at once and without further difficulty be able to determine the precise words that were read by them as Scripture, and a very strong presumption would be given in their behalf. Unfortunately neither of these two conditions is complied with.

As to the first, we know that the text of the New Testament, far from having continued pure during the first 3 centuries, had fallen into a state of remarkable confusion long before that period expired.

Several circumstances combine to show this:

(1) Manuscripts Disagree among Themselves: The differences among our ancient witnesses themselves, a point as yet only alluded to, and to be immediately spoken of more fully, are a clear proof of the fact. Had the text been preserved in its original condition they could not all at once have exhibited the diversity that we actually meet with. It was one at the first ; it would have been one then.

(2) Early Versions Disagree: Again, although we have no manuscripts older than the early 4th century, we have translations of the NT made into other languages nearly two centuries earlier, together with quotations from it embodied in writings of the Fathers belonging to as remote an age; and these translations and quotations leave no doubt that, when they were made, very many passages were read in their particular district quite otherwise than we find them in our oldest manuscripts.

(3) Early Fathers Confirm Existance of Variants: Still further, we can see from the works of the early Fathers of the Church that they were often greatly perplexed by the variety of readings that came under their notice. They speak of it continually, and often in tones of much hesitation and doubt. They refer to manuscripts even then older than others. They describe some manuscripts as being more accurate than others. They blame opponents for wilfully falsifying the text. They discuss the probabilities of different claims. In short, they find themselves largely compelled to pursue the same course of argument pursued by Biblical critics at the present day.

(4) Early Experts Discuss Variants: Finally, we have the express testimony of some of the most learned of their number, of some who devoted much pains to the study of the text upon this very point. We shall refer only to two, the most learned, the most critical of them all, one of whom flourished in the third century, the other in the fourth.

The first, Origen, commenting upon Matt. 19:19, where words occur falsely thought by him to be corrupt, says,

The second, Jerome, says, in his epistle to Pope Damasus,

It is true that Jerome is here speaking of the Latin texts, but the simple fact that they were thus corrupt is evidence enough that the variations in different manuscripts of the original must have been great.

Causes of Corruption of the Text

To what causes this corruption of the text about the beginning of the 4th century is to be ascribed it would be extremely difficult to say. Copying errors undoubtedly contributed to it, but they are not of themselves, by any means, a sufficient explanation. The fact, however, without going into this, is sufficient for our present purpose. It evidently presents a very serious difficulty when we would proceed to estimate the merits of our ancient witnesses who belong to that age.

The second difficulty alluded to is not less great. Our ancient witnesses are very far indeed from being unanimous in the testimony given by them. They are constantly at variance with one another.

We take one or two of them that seem to approach most closely; and for brief spans, we find agreement. Then they diverge. First the one and then the other joins a different group. But attachment to the new group does not continue long. Divergence from them speedily appears. As they go on in ever-varying combinations, it appears a well-grounded text is impossible.

And so it would be, if all our ancient authorities equally reliable. But its a mistake than to think so. Two manuscripts of the 4th century may differ from each other in worth quite as much as two of the 16th.

All the main causes that operate in bringing about a distinction in the latter operate also in the former case. The old were copies like the young. They may have been copied from bad originals. They may have been carelessly copied. Age alone cannot be accepted as decisive of their value. So we are back where we started. The two conditions that would have saved us effort do not hold. The only remaining course is for relative value and weight be attached to them.

Here one of the most interesting and important problems connected with our inquiry opens on our view. How do we decide between equally ancient but conflicting MSS? We shall explain the principle of the process.

It is important that we also have in early Versions (translations), and especially in early patristic quotations, evidence for readings much older than the oldest existing manuscripts.

It is true that this evidence is not free from the same influences as the MSS themselves. The text of translations and of quotations has been affected by time and contamination.

And published editions of these texts are also sometimes unreliable and their readings uncertain, for they too were once hand-copied. Still, allowing for such factors, the probable reading of many important texts used by early writers can be determined. But other resources and methods are also available.

Arguments for readings can be drawn from the general evaluation of their manuscript authority; for MS witnesses later than early versions and quotations can attest to earlier texts.

Internal evidence:

Internal evidence too, such as the context of the passage, the style of the writer, or the analogy of Scripture, may be taken into account. Each of these has some weight, and is part of the method by the 1st or 2nd century text is determined.

We establish the most ancient and certain readings then; and continuing as described in the last chapter, only now with much greater attention to detail, we test by means of these known readings the value of each ancient MS. If it contains them in the form we expect it is a proof that it is good. If not we rate the MS inferior.

It will be at once seen that the principle now advocated is thoroughly sound, not only to determine particular readings, but also to evaluate manuscripts as a whole.

Upon the first of these two points there can be no doubt. If we used all available evidence in settling the reading(s) of our characteristic texts, then a single dissenting manuscript (i.e., a singular reading) will not be surprising or create a problem. And whenever a MS does agree, it will only confirm our conclusion about the reading.

Extending Quality of Reading Back to Quality of MS

But the second point is equally indisputable; for surely it will not be denied that agreement with an established reading is a fair test of the value of a manuscript, in general.

Proven veracity in a witness upon many points is a reason why we should not only believe him upon these points, but why we should accept him as a generally credible witness on other points.

To refuse this principle, and a fundamental law of evidence is overthrown.

---

V: Third Stage - Part B

Using Readings to Evaluate MSS (p.109 fwd)

We have explained the principle by which the relative value of ancient manuscripts, all possessing great claims on our regard, is to be tested, and we proceed now to give one or two illustrations of the process.

(1) Matthew: 6:4

Let us take first a test to which reference has been already made, Matt. 6:4, " And thy Father, which seeth in secret, shall reward thee openly."

The word "openly" is found in several ancient manuscripts, but it is omitted in three of the most ancient. It is evidently necessary that we should know whether the former or the latter are most worthy of our confidence before we venture to decide between them. Now in connection with this text it happens that we have the means of doing so.

One of the most celebrated of the Fathers directed his particular attention to the question whether the word "openly " should stand in the text or not, and at last excluded it on the ground that, though found in many Latin translations of his day, it was not found in the Greek manuscripts, which were earlier in date.

This testimony is very strong. We turn to the most ancient Greek manuscripts that have come down to us, and the word is not found in them. It is also wanting in two or three important translations, and in citations of the text given by several of the Fathers.

Farther, as already pointed out in a previous chapter, the insertion of the word mars the sense of the passage, and leads to a different idea from that which our Lord evidently intended to convey. In these circumstances it is an easy conclusion from the evidence that "openly " ought to be omitted. That, however, is not what we have at present in view.

We rather lay aside one of the ancient manuscripts testifying to the omission. Our verdict on this point will not be affected. There is abundant evidence without it. We determine fully that the omission should be made. We then take up the old manuscript laid aside, and as yet supposed to be unexamined. Has it, or has it not, the word ? It wants it. Thus we have a proof so far, though one text will of course go but a little way in such an inquiry, that that manuscript is right, and more worthy of being relied on than any manuscript, even though an ancient one, in which the word occurs.

(2) Matthew: 5:22

Again, we take another very interesting text, Matt. 5:22. The reading of our English Bibles is,

Now we know from the express testimony of the two most celebrated critical Fathers of the Early Church, that the words " without a cause" were not found in the best manuscripts then existing, and that in their opinion, therefore, they ought to be excluded. Says one of them,

The other says,

As before, this testimony that the phrase should be omitted is very strong. With all their varied learning and great opportunities, these Fathers had considered the matter, and come to a clear and unhesitating decision.

The phrase is also wanting in two of our oldest manuscripts, in several valuable translations, and in others of the Fathers besides the two now cited. Internal evidence is also against it. So soon as we have reason to doubt the propriety of its presence, we see that it has all the appearance of a word inserted to avoid the apparent harshness of the precept without it ; while, if it was an original portion of Scripture, it is very difficult to give satisfactory reasons for the removal, the difficulty of the passage being thereby unquestionably increased.

On the other hand, the precept, if we omit the phrase, is in striking harmony with the at first sight sharp, extreme, almost paradoxical character of various other precepts of the Sermon on the Mount.

Although, therefore, we do find the phrase in a good many uncials, including in this case one of the oldest, in the great mass of cursives, in several versions and Fathers, it is not difficult to decide that it ought to go out of the text.

As before, however, that is not what we have at present in view. We rather lay aside one of the ancient manuscripts testifying to the omission, if possible, and it is here possible, the same one as before. Our verdict on the point of omission will not be affected. There is abundant evidence without it.

We determine fully that the omission should be made. We then take up the old manuscript laid aside, and as yet supposed to be unexamined, has it, or has it not, the reading? It wants them. Thus we have again so far a proof that that manuscript is right, and worthy of being relied on.

(3) Matthew: 28:9

We take still another text from the Gospel of St. Matt. 28:9,

The words here, "as they went to tell His disciples," are found in some Uncial manuscripts, but not in others. We desire to know upon which of these we may most rely. We pursue therefore exactly the same process as before:

Several of the Fathers bear distinct witness to the fact that these words should not be there, and so do two of our most ancient manuscripts (the one that we are testing being in the meanwhile laid aside), together with many other important authorities.

So far therefore as that point is concerned we can without difficulty come to a conclusion. The words have no right to their position in the text. We turn up the manuscript under examination. Does it have, or does it want, them ? It wants them. The conclusion as to its trustworthiness drawn in the two previous instances must be drawn again.

(4) John: 6:11

Once more, we take a very interesting passage from the Gospel of St. Jn. 6:11. It occurs in the account given by that Apostle of the multiplying of the bread,

Such is the reading of some of our authorities ; but others, omitting a portion of the sentence, read, "And Jesus took the loaves, and when He had given thanks. He distributed to them that were set down." The difference, it will be observed, is that, according to the latter reading, Jesus directly distributes the bread Himself; that, according to the former. He does it by means of His disciples. Now we find that all the most important versions, and all the most trustworthy Fathers who allude to the verse, so decidedly support the first of these two views, that we can have no hesitation in adopting it. The verse ought to read, "He distributed to them that were set down."

We return to the particular manuscript we are testing. How does the case stand with it? It wants the words ; and we have fresh confirmation of its value.

(5) 1st Cor: 6:20

One other example taken from the Epistles may fitly close this list. We take 1 Cor. 6:20, " Therefore glorify God in your body, and in your spirit, which are God's."

We can establish without the use of the manuscript with which we are dealing, that all the words here met with after "body" have no right to a place in the text, which ought to read simply, "therefore glorify God in your body." We turn to our manuscript as before. It wants the words, and we are strengthened in our old conclusion.

Summary

The same process has to be applied to each of our ancient manuscripts in succession, testing each by the mode in which it reads texts whose true reading we have been able to ascertain without it. In proportion as it agrees with them its value is enhanced ; in proportion as it differs from them it is diminished. In the five texts now selected one manuscript was found to stand the test of each. It is certainly more valuable, so far as five texts can help us to a conclusion, than one that could stand the test only twice, or thrice, or not at all.

The illustrations now given ought sufficiently to explain the general nature of the method resorted to for the purpose of determining the point that we have had before us. How we are to arrange and classify those ancient manuscripts that unfortunately so often differ from one another.

Our illustrations convey of course no idea of the very large number of texts to which the process must be applied, or of the complications that have often to be met before a final decision can be come to.

Yet enough ought to have been said to show that, when critics alter the standard text they don't do it at random. They know what they are about. By long, laborious, anxious study they have been able to establish certain principles by which they can decide as to the character of the witnesses before them; and they are thus prepared for giving their verdict upon the whole case in the calm judicial spirit of a judge upon the bench.

The effect also of the procedure now advocated upon the mass of our materials for judging of the true text of the NT will be at once apparent. It is a fact admitting of no contradiction, that the number of ancient manuscripts capable of passing the test is but small.

We have to add to them indeed those of our later manuscripts that pass with equal credit through the trial, for we have already seen that in such an event the modem manuscript is, as having an ancient text, to be ranked for our purpose along with the ancients themselves.

Again, however, this number is not great. The older and later MSS together which have the highest rating are few in number. Our circle of primary authorities has been still more narrowed.

---

VI: Results of Classifying MSS

Majority Text Dismissed, Bias Denied (p. 116 fwd)

Before proceeding further, it will be well for us to pause for a moment, and to consider the result that we have reached.

We have been endeavouring so to classify our witnesses to the text of the NT that we may distinguish their various merits, not putting any of them absolutely aside, but arranging them according to what seem their different degrees of credibility. We have been doing this too with the view of narrowing, if possible, our circle of primary authorities, so that we may not lose ourselves in what would otherwise be a labyrinth of confusion, through whose windings we have no clue to guide us.

(1) Ranking by Type of witness, Age of text:

We have seen that versions and quotations are less valuable than manuscripts of the text; that manuscripts possessing an ancient text have a greater claim on our attention than those whose text is comparatively modern; although if a modern manuscript possess, as some of them actually do, an ancient type of text, the mere fact of its being modern in form does not detract from its value. For the purpose that we have in hand it is really ancient, and must take its place along with those that are so both in form and in substance.

(2) Ranking Readings by Diversity of Independent Attestation:

We have also seen that we are in possession of a principle by which we can determine that certain even of the ancients are better witnesses than others, because they have preserved for us readings that we have independent authority for believing to have existed at the nearest date to the time of the Apostles, as to which we have any evidence at all.

(3) Reduction of Evidence to Primary Witnesses:

The result of the principles now advocated is so important, and at the same time so unexpected, that it is necessary to pause over it for a moment, and to see whether, in the presence of that result, we are prepared to abide by our conclusion. The result is that, notwithstanding the enormous mass of evidence that we have in our hands as to the text of the NT, our primary authorities are reduced to a very small number. It is an admitted fact, in particular with regard to manuscripts on which, as we have seen, our dependence is mainly to be placed, that by far the larger part of them do present a text differing to no inconsiderable degree, from that found in the few to which we have urged that a decided preference should be given.

(4) Objections to Dismissing the Massive Majority of MSS:

Formidable objection is therefore taken to this result. It is pleaded that the resolution to follow it up in the actual construction of the text is unfair to the great numerical preponderance of witnesses on the other side. It is conceded that character, not number, should prevail ; but the number against our few is, in the present instance, so very considerable that it is supposed to require a certain modification of a principle usually sound. There is against us a unanimity so extensive and so long continued that, when it is described even in language against which no charge of exaggeration can be brought, it is apt to leave an almost ineffaceable impression on many a mind that we ought to defer to it more than we have done.

(5) Quality of Witnesses Overrides Number:

If we reflect upon the matter for an instant we shall see that the idea thus entertained is distinctly to be repudiated. It is true that our witnesses stand by hundreds on the one side ; by tens, or rather units, on the other. If then we have reason to believe that the former are as good as the latter, their evidence will probably be as surely it ought to be, conclusive. Numbers, in such a case, it would be entirely out of the question to disregard.

But it can hardly be said that the presence of numbers, however great, makes us independent of investigating quality. In no inquiry that we can engage in is the character of a witness exposed to so many deteriorating influences as here. We must satisfy ourselves therefore whether each, as it comes before us, has suffered in this way or not. The mere assertion of any individual among them taken by itself is nothing, and multiply nothing by hundreds we have still nothing.

(6) The Onus to Prove Majority Text
    is a Corrupt Offshoot of Ancient Text is Denied:

It may be urged indeed that we are entitled to speak in this way only on condition that we can show that this majority of our witnesses have no independent character, that they gathered their information from the few, that they corrupted it in the process, and that we can learn all we wish to know from the lips of those to whom they in the first instance deferred. No obligation lies on us to show anything of the kind. We are by no means called upon to prove that these witnesses, who come to us in crowds, derived their information from the small number to whom we are disposed to attach supreme importance.

Their value as witnesses may have been affected by many other circumstances besides that. There may have been collusion among them. Without deliberate collusion they may have been led, under the pressure of the same powerful authority, to the same utterances. They may have all sprung from some one region of the world in which, owing to peculiar circumstances, some one, and that an imperfect, form of the text had gained supremacy. These things are at least possible. Whether they have actually happened or not demands inquiry. We are entitled to say that the credibility of all our witnesses must be tried by tests which every judge applies. If they stand the test they must be listened to, but escape it they cannot.

It is a matter of no weight whatever then in the present question that it so happens that, by the process we have been pursuing, the great majority of our manuscripts have ultimately to rank lower than the few.

(7) Bias of Textual Critics is Denied:

The textual critic is not to blame that it should be so. When he begins his labours he has no preference for an ancient over a modern, or for one ancient over another.

His first impression would probably be rather in favour of the great majority of modern manuscripts. He has been accustomed to the readings which they supply. He finds them easier, smoother, less perplexing in the forms and constructions of their words.

But the determination of the text of the NT is far too important, far too sacred a thing, to permit him to rest in what is familiar and easy. He must ascertain to the best of his ability what is true. Therefore it is that he is bound to test the value of every witness who comes before him, of all the evidence that is offered him. Whatever his own predilections might be he must follow the course suggested by reason and experience as intimate, knowing well that, though the result may not be what he himself would wish,

---

VII: Grouping Witnesses

(p. 121v fwd)

Up to this point we have been speaking almost wholly of manuscripts of the Greek NT, witnesses that we saw at the outset we must mainly rely upon.

We have found that we can classify them, and that we can thus introduce order where at first there seemed to be nothing but confusion. Can we now simply arrange these manuscripts according to plan, and take our text from the best of them? If they the only authorities to consider, and they took us back to the time of the sacred autographs, it would be that simple.

But neither of these is the case. We have already noted NT translations and patristic quotations older than any known manuscript ; and noted that our manuscripts reach only as far as the early 4th century; And also, that before that was a period when there was no small confusion of the text.

We must then recall other witnesses; for, though they are not primary authorities, they cannot be neglected. Won't this re-introduce, at least to a large extent, the old confusion which we persuaded ourselves we had escaped? At first it seems so.

But the phenomenon is far too interesting and important to allow discouragement; and we must seek something else to help us in evaluating our evidence in all its parts.

We turn back then to early 4th century variants, and we note that some groups of readings appear to have been prevalent in some geographical areas than in others.

There is by no means unmingled confusion, or a total lack of order in the variants; but patterns in the distribution.

Similarity between text-types in Constantinople and Antioch:

Thus, for example, it is found that in Gaul, Italy, and Africa there is a type of variation seemingly different from that which prevails at Alexandria or Constantinople; that at Constantinople there is a text-type markedly similar to that of the Fathers at Antioch in Syria.

If the facts bear out the correctness of this impression, it must have an important bearing on our inquiry, because it is clear that, given two contending readings of a text, the reading having the best claim to being original will be the one which held sway in the greatest number of locales, notwithstanding the tendency of these locales to add changes of their own.

Diversity of Geographical Attestation and Groups:

A reading's permanence amidst shifting ground shows its vitality and power ; and even if it has not been accepted everywhere, the more widespread the diffusion, in other words, the greater the permanence, the greater the power. We must look then at this matter somewhat more closely. It may be described as the principle of grouping.

Had the copies of the sacred autographs been confined to one district there would probably have been no scope for this, and no need to make any such attempt.

Special Nature of Versions (translations):

The materials and the need of grouping both arise when these copies are taken into other lands, are transcribed there, and are dispersed, partly it may be in the original, but mainly in translations. The inhabitants of one country differ from those of another in thought, in taste, in manner of expression, in occasional choice of words, even when their language is the same.

These differences are strengthened when a different language is spoken. The words of one tongue often fail to cover exactly the same field of thought as those of another ; and when we have to reason backwards from a Syriac, an Egyptian, or a Latin translation to a Greek sentence, it may well happen that we shall not always be led to precisely the same result.

This, however, is not all. The circumstances amidst which the work of transcription is undertaken in different countries also differ. In one there may be but a small demand, and the work may be gone about leisurely, slowly, and in a scholarly spirit. In another the demand may be much larger, and there may be greater haste and imperfection in its supply.

There may be no bad faith. All may be done in the strictest honour, and with the most sincere desire to produce faithful copies or accurate translations of the original.

The influences leading to change may work quite unconsciously in the minds of the transcribers or translators. The important fact is that they are there ; and that it is as impossible to be altogether free from them as it is for a succession of scribes in the same country to exhibit a perfect immunity from those errors of transcription, due to human frailty, that were formerly taken notice of and illustrated.

The effect is obvious. In different districts of the world different groups of errors will become prevalent, and the manuscripts, versions, or quotations containing the texts erroneously copied will present a certain family resemblance to one another.

Example: John: 2:3

Thus, for example, in Jn. 2:3, in the account of the miracle at Cana of Galilee, we read, "And when they wanted wine," or, as it might be more simply and literally translated, "And when wine failed, the mother of Jesus saith unto him, They have no wine."

In several of our important authorities we find, instead of these words, the following, "And they had no wine, because the wine of the marriage feast was finished. Then the mother of Jesus saith unto him, They have no wine."

We do not so much ask at present which of these two readings deserves the preference. We observe rather that the authority for the latter is, with the exception of one Greek manuscript, wholly Latin. It was mainly confined to those parts of the world where the Latin tongue was spoken. It had no hold either of the Syriac or the Greek East.

Not that groupings can be strictly carried through all our authorities as to readings. We cannot definitely assign them all to particular families.

MS Travel and Mixture of Readings

Neither manuscripts nor translations were confined to any one region of the Church. They passed from country to country and from city to city through that interchange of letters, books, and visits which forms one of the most interesting characteristics of the early Christians.

Mixed manuscripts thus came into existence - manuscripts showing traces of different districts, and bridging over the gulf that would otherwise have separated them. Yet in very many cases the lines of demarcation are sufficiently distinct to entitle the critic to speak of different styles of text corresponding to different regions of the Church.

Let us proceed to the bearing of what has now been said. It is highly important in the following respects.

1. Groups represent Streams: In the first place, mere number of manuscripts belonging to any group is of comparatively little importance; because, whatever the number of such manuscripts, they mainly testify to the one archetype, the master-copy at the head of the local family from which they sprang, and that head-copy of the family (group) can only count as one.

Wherever then we have a group of authorities, that group must count as only one stream of evidence, without reference to the number of individuals of which it is composed.

2. Counting Streams: In the second place, the greater the number of streams of evidence flowing from different quarters of the world and testifying to the same reading, the more likely is that reading to be correct. This is something entirely different from the number of individual manuscripts, of which one stream may exhibit many more than another.

The evidence of all flowing in each stream is taken as a whole ; and then the more streams we have of the same character the greater the confidence with which we infer that the reading floating on their surface comes from the head fountain of the waters. It is clear that it must be so. How do we know that any reading is absolutely correct? By the confluence of all our streams.

Example: John: 1:1

What is it that assures us that this sublime opening of the Gospel of St. John really came from the pen of the beloved disciple? That it was everywhere and at all times read. In every region of the earth to which a manuscript of the Gospel was taken these words were copied from it exactly as they stand. Gaul, Italy, Africa, Alexandria, Palestine, Syria, Constantinople, all read them as we read them now. Many another text in the same Gospel underwent great and different changes. These words underwent none; and the only possible explanation is, that they existed in the Apostle's autograph.

The same principle leads to the just preference of two agreeing streams of evidence to one, for that agreement is a testimony to the fact that, at a date anterior to the decay and corruption of the text, the reading given was more widely spread abroad than any other with which it has to contend. The only probable explanation again of this diffusion is that the reading came from the original, and possessed sufficient power to resist the influences that were producing change.

Example: John: 1:51 / Matt: 26:64

Let us illustrate what has been said by a reading in Jn. 1:51. There we find the Saviour saying, in our English version,

The words "From henceforth" are found in one of the leading Greek manuscripts, in several other uncials, in the whole family of manuscripts of a later date, whose headquarters were the Byzantine Empire, in two Latin versions of a later type than the earliest, in the Syriac, and in two or three of the later Fathers.

At the best, therefore, if read at all in early times, they were read only in Syria, and in Conittantinople, which was dependent upon it.

On the other hand, the words are omitted in three of the oldest and most valuable Greek manuscripts, in all the best Latin versions, in the version of Lower Egypt, in the Armenian and Aethiopic versions, and in some of the most valuable patristic writings that we possess. We have thus one region of the Church alone against all the rest, and that the region where we know from other evidence that manuscripts underwent most change in the hands of hasty, if not imperfectly informed scribes.

There can be no doubt that the reading of the verse without them is the true reading of what the Saviour said; although, if number of individual manuscripts were to be considered, the majority in their favour would be overwhelming.

It is worthwhile to observe that, in this case, we can discover without much difficulty how the words in question got their way into the text. The scribe had in his mind the language of Jesus in Matt. 26:64,

The two verses have a great similarity of tone. "From henceforth" was met with in the last and most familiar one ; it easily slipped into the first and less familiar.

We add only, in conclusion, upon this point, that it does not matter, though, as in the example now given, all the individuals in each of the groups which harmonize upon the whole do not bear witness to what we accept as true.

That so many of them do agree has to be accounted for, and the only explanation being that in each case they had drawn the reading from some common progenitor, the fact that these progenitors in different regions of the world had, at a much earlier date, borne the same testimony, is a presumption of the strongest kind that they correctly represent the one original.

---

VIII: Choosing the Text - Part A

External Evidence (p. 129 fwd)

Our witnesses are now classified and grouped, and we have to turn to the interesting and important task, to which all that has been hitherto said has been preparatory, - the hearing of the evidence, and the determination, by means of it, of those words in which we are to find the light of life.

At the point at which we stand it is to be observed, that we, suppose ourselves to be absolutely ignorant of the words of the New Testament. We are not in the position of one who has that volume in his hands, and who, having heard that the reading of some of its passages is disputed, is about to examine a number of witnesses as to what they have to say upon that point.

For more than two centuries after the Reformation, indeed down to the third decade of the present century, this was the position of successive editors of the New Testament. The Received Text, as it was called, had acquired such authority, that the utmost they ventured to attempt was to make such emendations upon it as seemed to be required by the ever-increasing knowledge of manuscripts and other sources of evidence. The greatest advance made by any of them was to ask, Is there reason to depart from the ordinary reading, and, if there was, to make that departure without hesitation ?

Lachmann's Abandonment of the Traditional Text

At the date, however, of which we have spoken, one of the most distinguished of the noble band of scholars who have devoted themselves to inquiries connected with our present subject, Professor Lachmann of Berlin, took another and decisive step. He started with the question, Is there any reason to depart, not from the Received Text, but from the readings that are best established?

His predecessors might have shown their modesty by adopting no new reading they were unable to defend. It escaped their notice, according to him, that it was unreasonable to admit any reading at all into the text, evidence of whose value they had not obtained. They, in short, had made the Received Text their point of departure.

He threw it wholly aside.

He would gather his whole NT text from the original witnesses, and from them alone.

This principle, undoubtedly just, is now universally accepted, and hence the interest and importance of the point that we have reached. From the witnesses whom we suppose to be before us, the whole text of our New Testament is to be gathered.

In estimating the value of their evidence certain principles or rules of judging must be taken into account. They may be conveniently divided into External and Internal.

---

I. Milligan's Principles of External Evidence.

1. The first and simplest principle by which we are to be guided is, that, where all our authorities agree, the evidence must be accepted as conclusive.

It is needless to enlarge upon this principle, which can admit of no dispute. Depending upon these authorities alone for our text, we have obviously no alternative but to accept what they with one voice proclaim to be correct. So much of the NT is guaranteed to us upon this principle that the disputed parts form only an exceedingly small portion of the whole.

2. Witnesses thoroughly tested in the manner previously explained are entitled to a hearing in every case equally respectful.

It may be true that an ancient manuscript possesses a certain advantage over a modern one in that, having been less frequently copied, it has probably escaped some of the changes which the mere act of copying is certain to introduce. Wherever, therefore, the departure of a recent manuscript from the reading of the older appears to be owing to frequency of transcription, the latter ought to be deferred to. But when the difference seems to have no connection with this source of error, when the variation is so distinct that it must be held to have belonged to the manuscript from which the more recent one was copied, it is impossible to allow that the greater age of an opponent shall alone decide the controversy in its favour.

We have tested the one as well as the other by the tests with which we started as the best and most trustworthy that could be devised. The one as well as the other has stood the test, and is therefore not to be undervalued because it may be written in cursive rather than uncial letters, or may be marked by numerals rather than the capital letters of the alphabet. The principle of the rule is evidently indisputable, but it is not unnecessary to point out that it ought to be adhered to. Any one consulting the critical apparatus, the lists of authorities, found in a critical edition of the NT, will soon learn that he is in constant danger of ascribing undue weight to manuscripts that have only a higher antiquity or a more imposing notation than that of many others to commend them.

3. With reliable manuscripts, quantity is an important consideration.

For it will be observed that we have not now number versus authority, the multitude against the "fit though few." By the supposition all have been proved credible witnesses, and although even then there will be many minute circumstances leading us to think one more credible than another, yet to put number out of view altogether were to neglect the principles of evidence upon which judicial questions are settled every day. But,

4. Mere number of witnesses in general, is of no weight whatever, without regard to the fact of their being or not being tested.

We have already seen that the coming forward of many in their numbers and with their unanimity may be the very best reason for rejecting them all without compunction.

5. The relative weight of manuscripts, versions, and citations must be duly observed.

However true it may be that by the latter two branches of evidence we test in no small degree the value of the first, it is not the less true that as a general branch of evidence the first is, when attested, entitled to the preference even over those by which it has been tried. The circumstance, however, that it is tried by them shows that there are certain cases in which they may claim to bear away the palm. Such cases must be carefully adverted to, and errors must be distinguished into their different classes, as for example of addition or omission, when we would know the special value attaching to each of our three great classes of authorities.

6. great importance is to be attached to the concurrence of authorities from different quarters of the world.

That is, to the meeting of different streams of evidence; to the combination of such groups as were previously described in favour of the same reading.

7. It is not to be forgotten that if any of these groups combine in favour of what we have reason to suppose to be the more ancient text, their verdict must be accepted as conclusive.

We have already seen that the simple fact that a reading is ancient does not prove it to be good. It may be a presumption in its favour, but that is all. But when, being thus ancient, it is supported by authorities from different quarters of the world; in other words, when a group of authorities unite in favour of what we can otherwise prove to be ancient, we have exactly the consideration added that was wanting to prove that it was not only ancient but correct. The wide diffusion of the ancient reading is established; that diffusion was owing to its vitality; and vitality is best explained by the supposition of originality and truth.

We have now pointed out what seem to us the most important principles of external evidence. Other writers have sometimes given others, or have arranged them in a different order. Indeed most critics who endeavour to edit a text of the NT have to some extent at least rules or principles of their own.

Before passing on it may be well to illustrate in a few sentences the application, in part at least, of the rules now laid down.

Example 1: 1st Jn: 5:7-8

We take a strong case first, the famous text regarding the three heavenly witnesses in 1 Jn. 5:7-8. As read in our English Version we find these words, a portion of which we shall enclose in brackets,

The question is, whether the words enclosed by us in brackets are genuine, or whether the text ought not rather simply to read, 'For there are three that bear record; the spirit, the water, and the blood, and these three agree in one.'

With the exception of two manuscripts, the one belonging to the 15th, and the other to the 16th century, the whole body of Greek manuscript evidence rejects the bracketed words in the above verses. These two manuscripts also do not stand the tests by which alone they can be shown to be worthy of much regard. Manuscript evidence could not be more decisive. We turn to versions. The suspicious words are wanting in every version except the Vulgate, nearly all the manuscripts of which exhibit them, those failing to do so being however generally recognized as the oldest and best.

Even where given, too, they are given with very considerable variations ; and such variations are always a proof of the doubtfulness of the authorities whence the words were taken. The evidence of versions is here as decided as that of Greek manuscripts. Lastly, we look at citations from the Fathers.

No Greek Father is known to quote the passage, while several of the most important, in arguing on the subject of the Trinity, to which it has so direct a relation, refer both to what precedes and to what follows, but do not make the slightest allusion to the disputed words. Nothing could more clearly show that they were unacquainted with them.

The most ancient and eminent Latin Fathers have likewise no knowledge of the words, and it is only in some of the later and more unimportant that they are discovered. The evidence of citations clearly accompanies that of manuscripts and versions.

In a case such as this it is obvious too, from what has been said, that all our streams of evidence combine, and that even in point of numbers the vast majority of the witnesses, including every one proved and tested, are on one side. The application of every rule or principle of External Evidence leads to the exclusion of the disputed words.

Example 2: Acts: 8:37

We take another case, Acts 8:37. In our English Version we read,

The question is. Whether these words ought to stand in the text, or to be excluded from it?

They are wanting in 4 out of the 5 chief witnesses described in the earlier part of this book: of the fifth nothing can be said, that portion of the manuscript having been unfortunately lost.

They are wanting also in a large number of other manuscripts. They are found in one uncial manuscript alone, although in a considerable number of cursives. The case is by no means so decided as the last; but our principles hardly admit of any other conclusion than one unfavourable to the presence of the words.

We turn to versions. The verse is wanting in the best codices of the Vulgate, in two Syriac versions, in the versions of both Upper and Lower Egypt, and in the Aethiopic. It is found in a codex of the Vulgate as well as in its printed text, in one of the Syriac versions, though there marked with an asterisk, and in the Armenian.

Our first conclusion unfavourable to the verse is thus so far confirmed. Lastly, we look at citations from the Fathers. The verse was known to one or two of the Latin Fathers, but we have positive assurance that it was not recognized in any of the writings of Greek Fathers that have reached us, until we come to one who flourished in the 11th, and another who belongs to the 12th century. That it should appear in the writings of the former need not surprise us, when we remember that the Latin versions are the chief authority for the words.

That it should be wanting in the writings of the latter is strong confirmatory evidence that it was unknown in Greek. Here again, therefore, the evidence of citations leads to the same conclusion as that of manuscripts and versions.

The argument however is not in the present instance quite so strong as in the last. We ask therefore with more interest than then, What is the situation with our groups of witnesses?

In one district of the world only does the verse appear to have been read, and that a district where we know from other evidence that there was a strong tendency to interpolate.

Our groups therefore deliver the same sentence as our individual witnesses. Finally, the number of tested witnesses against the words is greater than that upon their side. We sum up the evidence as a whole, and there can be no doubt that the verdict ought to be for the exclusion of the verse.

---

IX: Choosing the Text - Part B

(p. 138 fwd)

Having considered those principles of External Evidence by which the various readings of Scripture are to be judged, it remains for us to turn our thoughts to those principles of Internal Evidence which are neither less necessary nor important.

II. Principles of Internal Evidence.

Were the External Evidence on behalf of a particular reading in every case complete and satisfactory, we should have little occasion to depend on anything but it. The contest would at once be settled by an overwhelming balance of testimony, either on the one side or on the other.

It is rare, however, to find this in any disputed reading of much interest or importance.

There is then almost always a decided difference of testimony. Weighty arguments on their right to have a place in the text are urged by different claimants, and the balance of external authority is not unfrequently nearly equal. The application of Internal Evidence thus becomes indispensable.

It is so in that ordinary administration of law, the processes of which, as we have seen, afford the best analogy to the course which the Biblical critic has to pursue. A judge can rarely, if ever, settle a dispute between two parties by external evidence alone. It is the mind that sees, and not the eye. It is the mind that hears, and not the ear ; and according to the light in which different assertions present themselves to the judge's mind will be the judgment that he fonps.

The probabilities of the case, and the internal coherence of the narrative, must always influence his decision ; and his verdict is to be viewed as the hypothesis that takes up and explains all the phenomena connected with the dispute.

It is true that this necessity of reasoning on probabilities may often degenerate into mere subjectivity or wilfulness, and that a judge may carry out some theory of his own in such a manner as to set at naught well-established facts ; but therein lies the highest trial of the judge's skill. Therein judicial tact, ability, genius prove their infinite superiority to mere mechanical administration. For ten men who can learn rules, and apply them with accuracy to a case before them, we may be thankful to find one who, not acting apart from rule, can yet stand superior to rule, and can mould, in the fire of his own genius, both the external facts and the internal probabilities into one harmonious whole.

It is the same in the criticism of the text of Scripture. External evidence is not only valuable ; it forms the very ground of our proceedings ; it sets us the facts which we are to judge. But these we must judge.

The danger to which we are exposed of giving way to prepossessions, to subjective feelings, must be met ; and in the establishing of sound general principles, in the cultivation of a sound mind, lies the critic's power.

This much at least is certain, that no editor of the Greek text of the NT, except one, who gave it distinctly to be understood that his aim was special and provisional in its character, has attempted to construct his text upon grounds of external authority alone. Over and above such grounds, the resort to internal evidence has been always found to be necessary.

To these principles of Internal Evidence then we now proceed. The more important are the following: -

---

The reason of the rule is obvious. By the supposition we have two or three different readings of a passage, with external evidence not sufficiently precise to remove all doubt as to which the preference is due. Our fii^st object must evidently be to determine as far as we can their history, in other words, to ask how they severally arose. In doing so it is reasonable to conclude that the reading by whose existence the origin of the others is most easily explained is the correct one.

Example 1: Jn: 1:37

Thus, for example In Jn. 1:37 we have three readings consisting in three different arrangements of the same words. One of these gives the translation of our English Version, "And the two disciples heard Him speak, and they followed Jesus."

Another gives the translation, "And His two disciples heard Him speak, and they followed Jesus."

The third puts the pronoun in such a position in the Greek that the meaning is ambiguous ; it might be either of the two just mentioned. Which of the three may be best regarded as the parent of the other two ?

Not the first ; for there was nothing to make a scribe naturally slip into the second, while the ambiguity of the third is precisely what he would avoid. Not the second, for with it before him a scribe would be under no temptation to change it to the first, and as before, the ambiguity of the third would prevent his thinking of it.

The third, however, at once meets the necessities of the case. It is ambiguous. One scribe therefore, who viewed it in the first light, put down the words in the order there given. Another, who viewed it in the second light, made the order correspond to his impression of the sense. The third then is most probably the true reading.

Example 2: 1st Tim: 3:16

Another illustration of this rule is taken by Dr. Davidson from the celebrated text in 1 Tim. 3:16, and it is so suitable to the purpose that we shall again use it here. Three readings meet us in the first clause of that verse, that of our English version,

(a) "Great is the mystery of godliness,
God was manifest in the flesh," (or rather, "was manifested in flesh;")

(b) "who was manifested in flesh;"

(c) "which was manifested in flesh."

The external evidence in favour of one of these rather than the others is no doubt strong, but it is not so overwhelming as to make us independent of the probabilities of the case. We ask then, How does our present rule apply?

Let us suppose that "God" was the original reading. It is difficult to imagine how, with a reading the sense of which is in strict conformity with the teaching of the NT, a scribe should think of substituting "who", which, it will be observed, is then a relative without an antecedent. It is still more difl&cult to imagine how the important word "God" could pass into the neuter relative "which". The word "mystery" no doubt supplies an antecedent in this latter case, but the change is too great to have been made inadvertently.

Again, let us suppose that "which" was the original, it could neither pass easily into "who" without an antecedent while itself has one, nor into "God", which involves a change of the greatest magnitude.

Finally, let us suppose that the original reading was "who" We see at once, even without taking into account some particulars connected with the style of writing which greatly strengthen the inference, how it could pass into "God" "God manifest in flesh" was known to be "the great mystery of godliness". It was most natural to express it. In the hands of another scribe "who" might pass with equal ease into "which". The change of one letter only was involved, the sense was not altered, an antecedent was obtained where the want of one could not fail to be felt. "Who" therefore, as giving rise most simply to the other two, is the true reading.

It is interesting to observe in connection with this case, that, after long and keen discussions, scholars are now generally agreed that even external evidence, when at least we depend mainly on those authorities whose high importance we have advocated in these pages, leads to the same conclusion. We have thus our confidence in the soundness of what has been said both as to them and as to our present rule much strengthened.

---

The critic Bengel, not less distinguished for his piety than his talents, who first suggested this rule, has himself said that he regards it as but one application of the far wider principle, that good is difficult and evil easy of attainment.

The reason of the rule is obvious. A scribe was far more likely to substitute an easy for a difficult than a difficult for an easy reading. In speaking in the earlier part of this volume of the causes of various readings, allusion was made to the point now before us. But if, as then explained, a scribe was apt in copying a manuscript to substitute a simple for a hard expression, the converse must hold good that, in restoring the true text in cases of dispute, the hard is entitled to the preference.

Let us illustrate what has been said by one or two simple examples of different classes belonging to the one general principle.

(a) A reading at first sight obscure is to be preferred to one that is plain and easily understood.

Thus, in Jn.7:39, we read in our English Version, "For the Holy Ghost was not yet given, because that Jesus was not yet glorified."

It will be observed that the word "given" is in italic letters, showing that it did not exist in the Greek text from which our version was taken, and many most important authorities present this as the true reading. Many others however add the Greek word for "given", and we have to decide between them. Our present rule comes to our aid. When we read without the "given" the text is much more difficult to understand than when we read with it. A scribe, therefore, was less likely to omit it if he found it in the text before him than to insert it if it was not there. The conclusion is that in the original it probably did not exist.

(b) A reading presenting a historical difficulty is to be preferred to one from which the difficulty is removed.

Thus, in Mark 2:26, we read in our English Bibles of David, "how he entered into the house of God in the days of Abiathar the high-priest."

Various important authorities, however, omit these last words altogether, while others, retaining them, substitute the word "priest" for "high-priest". Both changes obviate the serious difficulty arising from a comparison of this passage with 1 Sam. 21:1, where the incident in David's life here referred to is related in the words, "Then came David to Nob, to Ahimelech the priest". Precisely on that ground, however, are these authorities to be suspected. A scribe knowing the difficulty would be much more ready to omit them if they had a place in his text than to insert them if they had not.

(c) A reading in one Gospel which seems to convey a sense different from that of a parallel passage in another Gospel is to be preferred to one which makes the two Gospels strictly harmonize.

Thus in Matt. 9:13 we read, "For I am not come to call the righteous, but sinners to repentance." The last two words, however, "to repentance" are frequently omitted. Shall we read them, or shall we not? Looking at the question for the present only in the light of Internal Evidence, we have to consider that the words are found in the parallel passage, Luke 5:37, where they are certainly genuine. The probability is, that from that text they found their way into St. Matthew.

---

We notice only one additional rule of Internal Evidence:

Thus in Jn. 13:24 we read in our Authorized Version, "Simon Peter therefore beckoned to him that he should ask who it should be of whom He spake." But there is much authority for Greek words that give us the translation, "Simon Peter therefore beckoneth to him, and saith to him. Tell us who it is of whom He speaketh."

The former reading requires the verb "should be" to be in the Greek optative mood; but St. John never uses the optative. It is not likely therefore that he would depart on the present occasion from his usual practice: and the last of the two readings mentioned thus finds corroborative evidence in its support.

Or, let us take an example in which mode of thought rather than merely literary style becomes the object of consideration.

Jn. 6:11, formerly considered in another aspect, will supply an illustration. The question there, it will be remembered, is as to the omission of certain words, the effect of omitting which is to represent Jesus Himself as distributing the bread to the multitude, while the common reading brings in the intervention of the disciples.

It is disputed which of these two readings is the right one. Let us bear in mind that the great object of S. John's Gospel is to set forth the glory of the Redeemer, to present Him to us in His single and unapproachable majesty as the Giver of all good, and we are at once led to conclude that the reading which favours this idea is most likely to be correct. It is needless to do more than say that either reading is equally consistent with the facts of the case. There is simply a slight difference in the point of view from which the writer speaks.

Other rules of Internal Evidence are given by different writers additional to those now mentioned. But even were they of more value than they are, we have said enough to convey to our readers an impression of the character of that part of the NT critic's work that is now before us.

It has only to be observed, in conclusion, regarding these rules, that each may easily be pushed too far, and may be used by the critic to reach his conclusion with too great rapidity.

It may well happen in many a case that the reading which seems most naturally to present itself to us as the parent of the others is not really so; that the plain reading may have been, by the ignorance of some scribe, transformed into the obscure instead of the obscure into the plain; that from the same cause a historical difficulty may have found its way into a copy when there was none in the original; that parallel passages which really correspond may have been brought to differ, instead of parallels Ihat differ having been brought to correspond; that a writer may at times use a method of expression different from his common one.

All these things are possible, and the too rigid application of any one rule might thus easily betray us. The lesson to be learned is one of caution, and that probabilities must be balanced one against another if we would hope to reach a conclusion on all sides capable of defence.

Finally, it has to be noted that rules such as those we have been considering must never be so applied as to overbear External Evidence.

They are not guides in all cases, but helps in cases, of difficulty. They aid in determining our scales to one side rather than another, when, but for them, the balance would be too equally poised or too uncertain. Even when not absolutely needed, they often confirm the verdict drawn forth by External Evidence alone; and, in doing so, leave no doubt upon our minds that we have the very words before us in which the Almighty revealed His will to man.

---

X: Summary

Consistent Principles, Personal Involvment

pST the two previous chapters we have considered the leading principles or rules by which we are to be guided in estimating the weight due to the evidence laid before us, with respect to disputed readings of the NT. One or two general remarks may fitly close the subject:

1. Such definite principles as those that have been laid down are absolutely essential to the right prosecution of the task of which we have been speaking.

No task can be named in which there is greater danger of hasty or haphazard judgments, of judgments founded upon prejudice or determined by inclination. And certainly none can be named in which we are more bound to avoid with the utmost care every influence of the kind, and to come to our conclusion only by steps, every one of which has been taken with the utmost deliberation, and in a way that we feel ourselves able to defend.

Our rules are a valuable aid in enabling us to do so. They are not arbitrary. They are founded upon a wide induction of particulars. They shape themselves according to reason and the nature of the case.

The knowledge of them thus directs us to a higher and more correct exercise of our powers than we could have made without them. We are taught to avoid many mistakes into which we might otherwise fall. We gain the feeling of security that belongs not to instinct but to intelligence ; and even where our rules are insufficient of themselves to lead us to our verdict, they indicate certain principles which the verdict, when we reach it, must not contravene.

2. They are not to be thought of only as individual rules, to be applied one at one time and another at another, in an outward and formal manner.

It may often happen, no doubt, that one is sufficient for our purpose. But the instances are far more frequent where many or even all of them must be applied.

They must be held therefore in combination. They must thoroughly penetrate and pervade the mind, so as to exercise a modifying and restraining influence upon each other, and to be ready for use, either singly or in ever-varying combinations, according to the nature of each particular case.

Nothing indeed will force itself more upon the mind of the textual critic of the NT, as he pursues his labours, than the impossibility of coming to a satisfactory conclusion in innumerable cases of dispute by the mere reckoning up of witnesses, or such mere calculations of probability as formal rules supply.

Every particular case needs to be considered in itself. Every witness needs to be weighed not simply as a whole but in the particular book in which the text under examination occurs.

Every suggestion arising from internal probabilities needs to be balanced by its opposite. The history of all the changes that have taken place on the original reading must as far as possible be discovered, so that that original reading, when decided on, shall contain in itself a solution of the problem, How what was one at the first came to assume so many varying forms in different countries and at the hands of successive scribes.

All this, it is evident, demands extensive knowledge, great calmness and impartiality of judgment, and the tact which, when not a happy natural endowment, can be gained only by long practical experience.

In point of fact, accordingly, successful editors of the text of the NT have been very few in number. Had the work been one of formal rule they would have been many; but the demands which it makes are so great that a comparatively small number have attempted it, and of these only one or two have gained a lasting fame.

3. No one ought to leave the matter in the hands of the few. The ordinary minister of the Gospel, the student of divinity, or even the cultivated private Christian, should not leave the whole matter only to those who can devote to it the whole labour of their lives, and feel that even that is too little to accomplish what they undertake.

Where we cannot discover we may yet follow with delight the discoveries of others. Where we cannot make original suggestions we can in a great degree estimate the value of suggestions that are made to us. Where we cannot combine the steps of an intricate proposition, we can judge of the accuracy of the combination set before us. We can analyze where we cannot create, and verify where we cannot produce.

In so doing, we share at least some part of the pleasure experienced by creative minds ; we enter into fellowship with them ; we learn to know the principles upon which they work; and we catch, although it may be afar off, some of the gleams of their inspiration. Nor is our imperfect knowledge without even a, wholesome influence upon them. It is a check to what might otherwise be their onesidedness. It spurs them to exertion and rewards their toil.

No mistake therefore can be greater than to think that, because the private Christian cannot do all in the work of textual criticism, he should do nothing. If, by studying its resources, its aims, and its principles, he gain no more than the conviction that it is really a science, the gain will be of the highest advantage both to himself and to the Church of Christ. With this confidence we now proceed to the third and last division of our subject.