to prove that during the Pauline period early Christianity had produced sporadic forms of epistolary literature, and at least the embryonic phases of what subsequently came to be wrought up into evangelic narratives. (ii.) Together with this feature, another must be reckoned. About twenty years elapsed between the crucifixion and the earliest of Paul's epistles. During this time, and even previous to his conversion, a Christian life was active, which did not owe its origin to him. He found churches in existence when he became a Christian, and alongside of his activity other agents worked more or less independently of his principles. These factors, and others like them, have to be taken into account in forming an adequate estimate of the period between 30 and 70. The accident that only Paul's theology survived in literary products, and that the minor contemporary currents failed to win any equal or at least immediate record, ought not to be allowed to distort the historical view into an undue exaggeration or depreciation. This is one of those cases where again it must be said that the written expression of an age needs to be corrected and supplemented by the recollection that the real importance of any movement is not to be adequately measured by the literary memorials which it afterwards secured. Before and through and round the Pauline letters, the mind's eye has to see much that cannot be set down in black and white.1

Facts like these bring out very forcibly the introductory and limited character of a chronological edition. To know the birthday of a book, as Dr. Martineau insists, is still a long way from a settlement of its parentage. A longer way,

On the relation of Paul's theology to the teaching of Jesus, there are very fine essays by Wendt (ZThK (1894), pp. 1-78) and Gloatz (SK (1895), 777–800). Generally speaking, we may say that in investigating the facts and beliefs that lie between 30 and 45 (or even 65), we are peering through a haze which renders their outline uncertain at many points, and occasionally prevents us from being sure whether we are viewing a given object in its true proportions, or whether indeed it is not an unsubstantial illusion. No contemporary documents exist. The main guide is inference based on later writings and developments, from which the historical imagination argues back with more or less penetration to the course of anterior events.

one might add, from decisive conclusions upon its value and trustworthiness. Certainty on the date does not win everything at once. The supreme adjective for chronology is "preliminary," and few will be so hasty as to imagine that, even were the question of dates more settled than it is at present, a corresponding assurance would have been thereby attained with regard to the historical contents and connections of the records. Their historicity and inner relations are always further problems, although it is upon these more than once that the question of the date partially depends. Consequently, while it is possible to tabulate luminously and honestly what seem to be results of thorough criticism— though provisional, they need not be indefinite-the attempt must be prefaced by the reminder that they do not form the whole, seldom even the major part, of the critical business. Beyond them lie the burning questions.1

For the most part a similarly provisional character attaches even to the "date" of a document. That also has to be taken in a somewhat loose sense. Usually it is equivalent to a circa of one or two years, occasionally to a larger period, during which the writing is first known to have been in circulation. Only in a few cases, like those of the Thessalonian and Corinthian letters, can the exact year, and even the month, be determined. The fact is, a consensus of opinion is to be

"

At the same time, exception must be taken to the unqualified remark that 'the doctrinal contents of an epistle may be correctly and adequately exhibited, whatever view be held respecting its author or its date" (Stevens, NTTh, p. 248). Surely, e.g., the epistle of James is one thing in the pre-Pauline period, and a very different thing in the post-Pauline. 1 Peter becomes in the seventh decade a writing of such spirit and significance as are considerably altered when it is taken some twenty or sixty years later. Change the locus of an epistle, adopt one view or another of its authorship, and the lights inevitably shift. In fact, the more accurately a writing is understood in connection with its age, the more vital to its interpretation are the problems of authorship and date. They seldom become altogether accessory or subordinate, nor are they in any case quite a matter of indifference either to the interpreter, or to any one who endeavours to use such a document carefully in his reconstruction of early Christianity. * The later ecclesiastical term δεδημοσιευμέναι γραφαί (Origen) suggests mainly the public reading of the writings in church (=publicari).

3 In dating the OT writings upon a similar scheme, the dialects and idioms of Hebrew are of large service (cp. Margoliouth, DB, iii. p. 33 f). A change in

gathered from the conflicting special investigations of scholarship only by some such self-denying ordinance of abstinence from minutiae. Fortunately, for most practical purposes it is not necessary to go further into details. As a rule the relative date of a writing is sufficient, i.e. its place in the general scheme before or after certain other books, previous or subsequent to some fixed point in history. More than this often cannot and need not be demanded. In NT criticism, as emphatically as elsewhere, the Aristotelian canon holds Be content with attaining so much precision and accuracy as the nature of the subject in hand allows. Generally, with the exception to which I have referred, it is a matter of small moment to know the exact month or even year in which a writing was composed, and the mere passion for a date, as for a theological label to a writing, is easily carried over the bounds of healthy scholarship. Beyond a certain point, absorption in such minutiae becomes a distraction. It is not impossible-witness certain lines of hypercriticism-to neglect the cedar of Lebanon through the amount of wasteful attention paid to the hyssop on the wall. The balance needs to be more correctly struck in many cases. In fact the purposes of interpretation are excellently served, as a rule, by assigning to the various writings of the NT and their dates a range which refuses to be unduly precise, and is content for the most part with ascertaining their relative order. One might almost declare with Dr. Bosanquet, that vocabulary and syntax can be felt as one passes, e.g., from the older narratives of Sam-Kings to Deutero-Isaiah and the memoirs of Ezra-Nehemiah; similarly, to the criticism of writings like Ecclesiastes, Daniel, and Esther, the linguistic evidence of Aramaisms proves at many points invaluable. Unfortunately this aid of language fails in the criticism of the NT almost entirely. Differences can be traced between the Greek of one writing and another, but the scale of the literature is too confined and the time too brief for such idiosyncrasies to afford reliable data towards determining the chronology of the writings. Hellenistic Greek, as employed in the NT, does not fall into periods. Its varied elements help to differentiate one group of writings from another; but Latinisms or Hebraisms seldom if ever yield any sure materials for fixing or even verifying the relative position of this book and that. The principles upon which stylistic features can be safely used as a criterion for the date or grouping of a writer's various compositions, are stated carefully by Zeller, Archiv für Geschichte der Philosophie, neue Folge, iv. 1. pp. 1-12, "Sprachstatistisches."

occasionally it is something at least "to know when they were not written." 1

Just as these limitations do not interfere with the genuine advantage and aim of a chronological order, neither is that order disqualified by the fact that the grounds upon which it rests are partially tentative. To some extent, it is true, criticism has cleared the area of debate and sensibly reduced the more extravagant theories. There are signs that the trouble of the documents at least is abating. But this does not apply to every point or side of the question. To write with anything like justice and accuracy upon the criticism of the NT, even in regard to the dates of its literature, one is often obliged to employ a staccato and chilling repetition 2 of "perhaps" and "probably "; while to take any line of one's own means opposition here and there to a more or less weighty body of critics. Several of the writings still abide our question. Indeed, in almost every department of research upon the beliefs and customs of the early Christian age, gaps are discovered, points between which no connection is easily visible, intermediate stages that must have once existed and cannot now be reconstructed with sureness, blanks in the course and sweep of life which only the historical imagination can be relied on to fill up. All this affects the arrangement of the literature. Such employment of surmise and hypo

1 "How to read the New Testament," Essays and Addresses (1891), p. 159. Cp. Rainy, The Bible and Criticism, pp. 14-23. Some of the more recent movements in criticism are occasionally described as a "retreat from the second century"; but this phrase needs considerable qualification, and certainly does not support the vague impression which seems to prevail in some circles, that to assign a document to the second century is to stamp it as second-rate. Such an idea is an unhistorical misapprehension. No evidence exists to prove that about the year 100 A.D. a night of unclean and inferior things descended upon early Christianity, when the "good things of the day" began "to droop and drowse."

Though this is often practised to quite a needless extent. It is useless to follow the first part of Cicero's well-known maxim for the historian-ne quid falsi dicere audeat-without adding courage to caution and proceeding-deinde ne quid veri non audeat. Much more is definite in NT criticism than is commonly allowed, and the affectation of reticence and hesitation is due as often to intellectual looseness or incapacity, as to a proper desire to be scrupulously fair and accurate in judgment.

thesis puts the literary problems on conjectural ground; it forbids robust and unambiguous statements, and frequently makes any approach to unanimity impossible. Still, this is a risk that has to be taken and will have to be taken with any arrangement and at any time. Besides, it must be added, recent movements in NT criticism have made such an attempt at a chronological order much more feasible than has hitherto been the case, by clearing up one or two difficulties to the verge of actual probability. The days are past when the beginning of knowledge in many quarters seems to have been contempt for Eusebius and his authorities. Tradition is being wonderfully, though far from entirely, rehabilitated, and that implies a wider province of common agreement1 upon the individual and relative positions of the NT writings. This is true even when one hesitates to accept in toto Harnack's seductive and exuberant vindication of tradition,2 or the particular theories which he applies to the NT writings. There can be no doubt that by this critical tendency, of which his famous

1 From the standpoint of an intelligent and dispassionate outsider, the late Mr. G. J. Romanes was on the whole justified in claiming that the outcome of the great battle upon the Christian texts had been, "impartially considered, a signal victory for Christianity." As he pointed out, " prior to the new [biblical] science, there was really no rational basis in thoughtful minds either for the date of any one of the NT books, or, consequently, for the historical truth of any one of the events narrated in them. But now all this kind of scepticism has been rendered obsolete" (Thoughts on Religion (1895), pp. 155, 156). At the same time, as the Notes and Appendix will show, there are several points at which the need is to follow up tracks of fresh inquiry rather than to halt in any final conclusions.

2 In the Vorrede to his "Chronologie" (1897). It is unnecessary to quote the well-known sentences, particularly as their foundation has been rather shaken by the subsequent discovery of compositions like the Coptic "Acts of Paul" (cp. Dr. Schmidt, the editor, in ThLz (1898), 316, and Harnack himself, ibid. (1897), 629). That a work of this kind should be accepted by subapostolic tradition does not tend to increase one's confidence in that tradition, and certainly warrants any cautious investigator in refusing to accept statements simply because they are current in the church by the time of Irenaeus. Tradition, as an accurate channel for the transmission of genuinely canonical literature, does not deserve the blank certificate which Harnack seems or seemed inclined to award it. Further, the standpoint of his scheme with regard to the NT literature cannot be said to be exactly representative, nor does it afford any adequate grounds for the belief that it implies a conservative reaction in NT criticism.