The Secret Gospel

The Secret Gospel

THE SECRET GOSPEL

i. The blessing of the children (candidates), 169 ii. The requirements for baptism (monotheism, ten commandments, renunciation), 169 in. The mnemonic ov8e Ις αγαθός el μη €ίς 6 θβός, 170 iv. The ten commandments, 171 v. βμβλόφας άντω ηγάττησεν αυτόν, 1 7 1 vi. The abandonment of property, 172 vii. The creed prophecied (prophecy of the passion and resurrection), 173 viii. The creed exemplified (the resurrection story), 173

After six days, 175 Nocturnal, 175 The sheet over the naked body, 1 75 The mystery of the kingdom of God, 178 f. The omission from canonical Mk. of the secret text, 184 g. The Carpocratian peculiarities, 185 h. The concluding sermon, 1 86 x.

The baptism (the nocturnal initiation), 174 4. Evidence for abbreviation at Mk. 10.46, 188

E. Conclusions, 192 I. Date, form, and affiliations The assumption that the letter was written by Clement entails the consequence, remarked upon above (commentary on 1.23), that the secret Gospel was not written by Clement,1 but was accepted by him — rather against his personal inclinations — anyone could suggest that Clement had written the secret Gospel. The alternatives are either your view or the hypothesis of a later person's writing the whole thing. If that is the case, I am inclined to think that it might be a job done with no specific tendency, but mystification for the sake of mystification. A curious instance is P. Oxy. 412, where the learned Julius was either duped or faking for faking's sake.

. . . Another possible point of comparison is the work of the people responsible for the Clementine Homilies and Recognitions. I think that they had no tendency, like the author or authors of the Grundschrift, and at the same time they were not seeking personal fame like Euhemerus."

P. Oxy. 412 contains the conclusion of the eighteenth book of the Kestoi of Julius Africanus, including a quotation of Odyssey XI. 34-43 and 48-51 expanded by insertion of a transitional passage and a magical incantation; the incantation is re-edited with commentary as PGM XXIII. Julius says the whole of the inserted material (29 verses) was to be found in MSS in Jerusalem (Aelia Capitolina) and at Nysa in Caria, and the first 13 verses of it in a MS in Rome, in the library at the Pantheon. Such a brief interpolation in the text of Homer is obviously something quite different from the sophisticated composition which confronts us if the letter and its quotations are taken as the work of a single forger.

The Clementine Homilies afford a better comparison, but the comparison tells against the argument, for they make no effort to imitate the style of the genuine Epistle of Clement·, nor do they set their pretendedly early material in a speciously later frame. The most serious objection, however, against any such argument is that it is unnecessary. Almost any work of ancient literature can be supposed a forgery

(cf. L. Wiener, Tacitus' Germania and Other Forgeries [Philadelphia, 1920], a work of great learning, or the attacks on Aristotle's Constitution of Athens referred to by von Fritz and Knapp, p. 4, to say nothing of the

THE SECRET GOSPEL

because he found it already accepted by that church in Alexandria to which he attached himself when he came to the city, probably about the year 175. (Julius

Africanus said that Clement was already a prominent figure in Alexandria during the reign of Commodus, 180-192; Routh, II. 307. Clement's canon was contrasted with that of the Alexandrian church by Harnack, Origin no.) It would seem likely that the church's acceptance of the secret Gospel antedated Clement's arrival by some considerable time; the composition of the secret Gospel of course antedated its acceptance. To allow twentyfive years for these two intervals and so put the com position back to 150 would not be implausible.

But Clement's letter indicates an earlier date. It says the secret Gospel was first written by Mark, then stolen and corrupted by Carpocrates.

Christians of Clement's party were always accusing their opponents of corrupting and misinterpreting the Scriptures (Williams, Alterations 31; Bauer, Rechtglaubigkeit

186). Presumably their opponents brought the same charges against them, and the charges of both sides were occasionally justified : all parties among the early Christians revised the texts of their Scriptures to meet their doctrinal needs — omitting em barrassing details and inserting words they thought desirable — and imposed on these texts interpretations which were often false (against Bludau, Schriftfalschungen, see

Williams, Alterations 25-53; Bauer, Leben 492-504 and passim; for Clement's own practice, Buri, Clemens io8ff). Consequently, there is no reason to doubt Clement's statement that Carpocrates "corrupted" the text of the secret Gospel. Nor is there any reason to doubt that Clement's own text of the Gospels had been— in the jargon of modern criticism — "adapted to the needs of the growing Church."

We saw above that the letter contradicts the published writings of Clement by admitting that the Carpocratians derived their doctrines from the secret Gospel.

This is a damaging concession by the writer and therefore most likely true. The specious contradiction between the letter's two statements, that Carpocrates mis interpreted the Gospel according to his doctrine and that the Carpocratians drew their doctrine from the Gospel, is merely a consequence of Clement's shifting from his own account of what happened to a sarcastic paraphrase of the Carpocratians' claims. His account was that Carpocrates corrupted and misinterpreted the Gospel according to his own doctrine. The Carpocratians' claim was: "Our doctrine is derived from this source." Clement paraphrases the claim, without bothering to deny it, because he has just declared the source polluted. That the Carpocratians did claim to derive their doctrine from the secret Mark is suggested by Celsus' reference to "the Harpocratians who follow Salome" (Origen, Contra Celsum V.62, with Chad wick's note, ad loc.). "The Harpocratians" are pretty certainly the Carpocratians (the assault on the Dead Sea documents by which S. Zeitlin has more recently made himself more ridiculous).

But the supposition of forgery must be justified by demonstration either that the style or content of the work contains elements not likely to have come from the alleged author, or that some known historical circumstances would have furnished a likely occasion for the forgery. In the case of the letter, no such demonstration seems possible, and the supposition therefore rests on nothing more than the feeling that this just cannot be genuine. That feeling may be correct — given Nock's knowledge of Greek and his amazing intuition, one hesitates even to doubt it — but it is not, by itself, conclusive.

THE SECRET GOSPEL

god Harpocrates also appears as " Carpocrates" ; Nock, review of Harder 221) and Mark is the only one of the canonical Gospels in which Salome appears. She appears again in the fragments of the secret Gospel quoted by Clement, and the Carpocratian text seems to have given her a considerably larger role than Clement's did (III. 1 4-1 7).

However, she also appeared in the Gospel according to the Egyptians , and Gelsus may have been referring to Carpocratian use of that text or some other now unknown to us.

On the other hand, the story of how Carpocrates got the secret Gospel (inspired by demons and using magic, he so enslaved a presbyter of the church in Alexandria, etc.) is evidently of the cock-and-bull species. No doubt the Carpocratians were else where accused of practicing magic (commentary on II. 4) and did practice it — the practice and the accusation seem to have been almost equally common in ancient

Christian circles. But the story reports that the magic was efficacious. And if this be excused as Clement's notion of what had happened, yet the basic facts reported are polemic: Carpocrates got the secret Gospel from a presbyter "of the church," i.e., of Clement's party. This claim is intended to prove Clement's party the original possessor. Therefore, though not incredible, it is suspect, being so strongly motivated that, even if it cannot be proved false, it cannot be accepted as true without further confirmation. Nevertheless, it is important because of what it does not say.

It does not say that the secret Gospel was introduced into the Carpocratian sect at some recent date. This is a charge Clement would have been happy to make had he known any excuse for it and might have made without excuse had he thought it would be believed. But his words in II. 7 (εξηγήσατο) suggest there was a commentary supposedly by Carpocrates on the secret Gospel, which would be further evidence that the sect had possessed it ever since Carpocrates' time. (In Irenaeus' time the Carpocratians did have an interpretation of writings which reported secret teachings of Jesus [Harvey, 1. 20. 3 = Stieren, 1. 25. 5, εξηγούνται]. Basilides, whom Eusebius thought a contemporary of Carpocrates, wrote twenty-four books of εξηγητικά "on the Gospel"; and Papias, probably an older contemporary, wrote five books of εξηγήσεις of " dominical sayings": Eus. HE III. 39.1 ; IV. 7. 7 and 9; Stahlin II. 284. 5, etc.) At all events, the important fact is Clement's admission that the Carpocratians have had the secret Gospel ever since the time of Carpocrates himself.

The date of Carpocrates will be discussed below in Chapter Four. He evidently worked in or before the time of Hadrian (117-138). Moreover, if he adapted the secret Gospel to his own purposes and represented it as the basis for his teachings, which Clement indicates he did, he must have got hold of it at an early stage in his career — at the latest, one would guess, before 125. Moreover, unless we suppose

Clement's church took its secret Gospel, for use in its "great mysteries," from the Carpocratians, we must suppose either that Carpocrates got it from Clement's church — as Clement says he did — or that both Carpocrates and Clement's church got it from some common source. In either event we shall have to suppose the secret

Gospel somewhat older than Carpocrates' adoption of it. Thus acceptance of the letter as Clement's entails admission of a probability that the secret Gospel described by the letter was in existence well before 125.

THE SECRET GOSPEL

Besides indicating this terminus ante quem for the secret Gospel, the letter gives us some notion of what this Gospel was like. First, it was a Gospel "according to Mark"

— this was the claim of both the Garpocratians (I.nf) and of Clement (I.2iff). It certainly included at least parts of the present canonical Gospel according to Mark : to such parts Clement gives precise references (II. 21, 22; III. 1 if, 14). It probably contained all of canonical Mk. — Clement says it was composed by additions to the canonical Gospel, but says nothing of omissions (I.2of,24ff). The additions, Clement says, were made by Mark himself, of material from his "notes" (υπομνήματα, I.i 90 and those of Peter. The new material did not exhaust these notes, but was chosen from them. It consisted of "things suitable to those studies which make for progress toward, knowledge " (τά τοΐς προκόπτουσι περί την γνώσιν κατάλληλα , I.20f), both of stories (πράξεις, 1. 24) like those in the canonical Gospel and sayings (λόγιά τιν a,

I.25) of which the exegesis would lead the hearers to the hidden truth (1. 26). It did not contain τά άπόρρητα (I.22f), nor "the hierophantic teaching of the Lord" (I.23—

24, probably identical with τά άπόρρητα) . The expanded text constituted a "more spiritual Gospel" (πνεύμα τικώτερον ευαγγελ ιον, 1.2 if), which was intended to be useful to those who were being "perfected" or "initiated" (τελειουμενοι, 1.22).

This text was kept secret by Clement's church in Alexandria and read only "to those being initiated ( μυουμενοι ) into the great mysteries" (II. 2). It was in the custody of the presbyters of the church, or they had had access to it, so that one of them had been able to secure an inferior (?) copy ( άπόγραφον ) for Carpocrates (II. 5-6). Cle ment himself either had a copy or knew the text by heart or had access to it; he could quote it verbatim to Theodore.

The Carpocratians also had a text, but it differed from that of Clement's church.

Clement perhaps ascribed some of the differences to errors of the original copy (this may be implied by pejorative connotations of άπόγραφον, 1 1. 6), but his words give the impression that he thought the more important differences due to additions he described as "most shameless lies" (άναιδεστατα φευσματα, II. 8— 9). The adjective may be intended to characterize them as obviously false, or obscene, or both ; γυμνός γυμνω in III. 1 3 suggests obscenity, but the obscenity may have originated in Cle ment's interpretation. What would a hostile interpreter have made of the rubric in Hippolytus, Apostolic Tradition (XXI.11): "And let them (the initiate and the presbyter who is baptizing him) stand in the water naked"? [A.D.N. thought "false" was right, not "obscene."] At any rate, the Carpocratian text was longer than

Clement's in at least two instances (III. 13 and 17) and in the latter of these two it contained a good deal of additional material (τά δε άλλα τά πολλά, III. 17 — unless this refers to a number of unspecified citations, which is unlikely in view of the parallel to III. 1 3 where τάλλα refers to additional material in the passage discussed).

I mentioned above Irenaeus' report that the Carpocratians had writings allegedly containing the secret teachings of Jesus, which they interpreted (Harvey, 1. 20. 3 =

Stieren, 1. 25. 5). Irenaeus says nothing of these works' being secret and writes as if he had seen them. However, he seems to have seen a good many "secret books of his adversaries, so his knowledge of these does not disprove their secrecy, which, given his description of the Carpocratian sect, is a priori likely. How it happened that some

THE SECRET GOSPEL

Carpocratians used their Gospel according to Mark in their argument with Theodore we do not know. Compare Clement's accusation είσί δ'οι λεγοντες είναι γνωστικοί τοΐς οικείοις φθονοΰσι μάλλον η τοΐς εκτός (III. 1 45· 1 6f) . This use of φθονεΐν for those who keep things secret appears also in the letter, 1. 27. Irenaeus (Harvey, 1. 20. 4 = Stieren,

I.25.6), followed by Epiphanius ( Panarion XXVII. 6), identifies those who call them selves "gnostics" as the Carpocratians, but elsewhere other identifications are given

[A.D.N. suggests that this passage in Clement may refer to "people who say they are gnostic in Clement's ideal sense"].

Most important in the information afforded by the letter is the fact that the Carpocratian text of the secret Gospel and the text of it used by Clement's church were basically the same. This was the most embarrassing fact the letter had to explain; therefore it is the least dubitable of the data. Moreover, Clement's admission is confirmed — the two texts differed from canonical Mk. at the same places and about the same things. Therefore, in spite of their differences, the two must have had

— as Clement said they had — a common original. Accordingly, the question of their differences from this common original and of its differences from the text of canonical

Mk. is a question of the history of the text of the NT. The letter's evidence shows that in Clement's time there were at least three forms of the text of Mk. : a short form

(preserved in our canonical text) and at least two longer forms (one the possession of Clement's church, another, of the Carpocratians). The longer forms differed con siderably from each other, but were both developments of a single, original, longer text which itself had differed considerably from the short one. There is reason to think that this longer text was in existence well before 125. How, then, did it come into existence and what was its relation to the short text which has survived? Was the longer text produced by expansion, or was the short text an abbreviation, or were both derived by different changes from some common original ?

These questions Clement has answered in his letter. He says the longer text was l expansion, produced in Alexandria, of the short text which had first been written

Rome. Both the expansion and the short text were the work of Mark, and so on, as stated above. The incredible element in this story is the claim that both texts were written by Mark. This claim is almost certainly false for the short text — i.e., the canonical Gospel — therefore it can hardly be true for the long one (Bultmann,

Geschichte 1-4, 362-376; against Taylor cf. my Comments and Nineham, Eyewitness).

Clement's credulity about apostolic authorship has already been noticed (above, 1. 19 etc.). The statements that Mark came to Alexandria and wrote the longer text there may be guesses to explain why the longer text was preserved (or, known to Clement) only in Alexandria. However, Mark's journey to Alexandria may have been reported by Papias (above, on 1. 15). The statement that the longer text was produced by Mark's expansion of the short one may, again, be a guess; or it may reflect local tradition, which merely imposed a famous name on an essentially correct report.

Apart from the reference to Mark, the story is not implausible. All Christian Scriptures at this time seem to have been kept secret from non-Christians (Tertullian,

De testimonio animae 1. 4), and the moment in the training of the catechumens when

THE SECRET GOSPEL

they were first (?) permitted to hear the Gospel was evidently an impressive one

(Hippolytus, Apostolic Tradition XX.2). Therefore if Clement's church had further

"great mysteries" beyond the primary initiation (1. 22; II. 2; cf. Origen, Contra

Celsum III. 59) it might have developed a special Gospel to be used in them. And Mk. would have lent itself to such development because of its well-known esoteric traits (Wrede, Messiasgeheimnis i46ff; cf. Wikgren, ΑΡΧΗ).

Accordingly, there is nothing improbable in Clement's report that the longer text of Mk. originated in Alexandria, by addition of material hinting at secret doctrines, and that Carpocrates then got hold of it and adapted it to his own purposes. But there is nothing improbable, either, in the notion that Clement's church should have made up the story of Markan expansion to cover its ignorance of the actual origin of the longer text, just as the story of Markan authorship was made up to cover ignorance of the actual origin of the short text. We have already noted the polemic motivation for the story that Carpocrates got his Gospel from Clement's church.

Invention of such stories is usually observant of probabilities; therefore only the credulous will find in these probable stories more than a possibility of truth. To one who looks for objective evidence the preservation of the longer text in Alexandria will seem an argument for its Alexandrian origin, but the weight given this argument will depend on a study of the relation of the longer text to the short one.

This relation is not unique. Of OT books, Jeremiah, Esther, Daniel, and EzraNehemiah; of uncanonical pseudepigrapha, the Testaments of the Twelve Patriarchs and II Enoch; of early Christian literature, the Ignatian epistles and the Didascalia

Apostolorum : these are only a few of the many texts which have survived in several redactions. Even more are proved by internal evidence to be either abbreviations or expansions of texts now lost. The literary procedures which produced these phenomena are admirably analyzed in Bickerman's Esther.

Of NT books, the text of Acts is preserved in two forms so different that many critics have thought one a deliberate revision of the other (Hatch, Text ioff). On the analogy of Acts, Blass suggested that the differences between the "Alexandrian" and the "western" texts of Mk. might indicate that there had been two editions of Mk. ( Acta 33). Other scholars have come to the same conclusion from study of the synoptic problem (recently Brown, Revision). But even if this conclusion were accepted it would not be directly relevant to the problem facing us, for the differences between the two editions thus postulated would be small, while the differences between the two texts of Mk. known to Clement were so great that his church treated the longer text as a different Gospel. The two texts must, therefore, have differed at least as much as would the two editions of John, for which evidence has been found in the preserved Gospel (Parker, Two Editions).

During the past century it was often thought that the present text of Mk. had been produced by extensive expansion of a shorter Gospel. The most famous pres entation of this theory was probably Wendling's Ur-Marcus\ the most recent known to me is Trocme's Formation (ιβρίΓ). Trocme's book appeared in 1963? when this chapter of the present work had been substantially completed; the agreements ob servable hereinafter are the results of independent consideration of the evidence and

THE SECRET GOSPEL

as such may have some evidential value. Fortunately, however, we need not rely on analysis of the present text of Mk. for evidence that the Gospel texts were often extensively revised. Two longer texts of Mk., generally believed to have been produced by expansion of the canonical text but differing so greatly that they are commonly treated as different Gospels, have been preserved — they are the canonical Gospels according to Matthew and Luke. These are generally supposed to date from the last quarter of the first century or the opening years of the second (Leipoldt, Geschichte

1. 108 ; Jiilicher-Fascher, 286-319). It is a likely guess that they owed their preservation primarily to their acceptance by particular churches — Lk. perhaps in Greece, Mt. in Syria (Harnack, Origin 68ff). The longer text of Mk. produced in Alexandria was not preserved, no doubt because of its intimate connection with the esoteric inter pretation of Christianity which seems to have dominated the churches of Egypt during most of the second century (Bauer, Rechtglaubigkeit 5 iff; cf. Hornschuh, Anfange 32off).

Esoteric practice limited this longer text to an inner circle and so prevented its attaining even in Egypt the sort of regional pre-eminence to which Mt. and Lk. may have owed their ultimate acceptance by the whole Church.

Since Mt. and Lk. are generally supposed to have been produced by an expansion of canonical Mk., Clement's account of the origin of the longer Alexandrian text is supported by analogy, and further analogies might be found in the yet further ex panded texts produced by Tatian and Theophilus of Antioch. The process had not stopped in Clement's day; he knew some "who alter the Gospels" and quoted with approval some of their alterations, which were expansions (II.266.25ff). Moreover, the same process has been thought to have produced canonical Mk. itself, which Dodd, Framework , has represented as the expansion of a primitive outline by addition of various pericopae — a theory still plausible in spite of the attacks on it by Nineham,

Order ; Robinson, Quest 48!!; and Trocme, Formation 23ff. But analogy is not con clusive evidence. In textual history abbreviations are no less common than expansions.

The Ignatian epistles, for instance, underwent both. So did the text of Mk. : even the expanded forms produced by Matthew and Luke show abbreviation; Matthew often condenses the Markan stories and omits some; Luke omits a large section of the text

(Mk. 6.44-8.26). The expanded form produced by Luke was abridged by Marcion and the abridgment was represented as the original text — a claim which still finds occasional defenders (Knox, Marcion). The expanded form produced by Matthew seems to have been abbreviated by the Ebionites (Hennecke-Schneemelcher, 100).

Another abbreviation is probably represented by the so-called "Fayyum fragment"

(ibid., 74). The canonical text of Mk. itself is often supposed to have been abbreviated at the end, if not elsewhere (Williams, Alterations 44f; Taylor, 610). Dionysius of Corinth, about 170, complained that his own letters in his own lifetime had suffered both deletion and interpolation; hence, he said, there was no reason to wonder that even the Scriptures of the Lord had been tampered with (Eus., HE IV.23.12).

Accordingly, the question whether the longer, Alexandrian text was an expansion, or the shorter, now canonical, text an abbreviation, will have to be considered care fully. At this point all to be said is that Clement's account of an expansion (except for its claim of Markan authorship) is consistent with the probabilities of the situation

THE SECRET GOSPEL

in Alexandria, with the history of early Christian literature in general, and in par ticular with the history of the canonical Markan text as indicated by the other synoptics.

Additional evidence in support of Clement's account could be brought from the fragments of the apocryphal Gospels, of which some show further modifications of the basic texts of the synoptics, although others may be (like the canonical Gospel accord ing to John?) wholly or partially independent compilations from cognate oral tradi tions (Mayeda, Leben-Jesu-Fragment, in spite of Benoit's review; Dodd, Historical

Tradition 328 n2; Hornschuh, Anfange I7ff; Hennecke-Schneemelcher, 34, 47ff, 57ff,

104). Fortunately, however, there is no need to rely on this material of which the interpretation is so uncertain. The Gospel described by Clement's letter was un questionably a variant form of Mk. In structure its closest analogues would seem to have been the other synoptic Gospels. By vocabulary, style, and content it is obviously connected with the synoptics rather than the apocryphal Gospels. Accordingly, the importance of the apocryphal Gospels to the following discussion is chiefly as evidence that from 75 to 125 the production of Gospels was not limited to the canonical four, and traditional material about Jesus was not limited to the many Gospels produced.

On the contrary, all the Gospels, canonical and apocryphal alike, are but partial representatives of an oral tradition which still outranked them in the time of Papias

(ca. 125; Eus, HE III. 39. 4) and lived on at least in isolated figures to the end of the second century. (Irenaeus, in Eus., HE V. 20.5-7, was still "chewing" this "cud."

See also Klijn, Survey 164-165.) This background of oral tradition is reflected not only by the apocryphal Gospels, but also by the agrapha in the apostolic fathers, of which Roster's study has led him to the conclusion that: "The source of the synoptic tradition . . . is . . . the community, which from its practical needs not only hands down and uses the synoptic material, but also recasts, transforms, and increases the material already available. Actually, moreover, this whole development is still far from completion at the time when our

Gospels are composed, i.e., toward the end of the first century. It continues, indeed, not only on the basis of the now developed Gospel texts, but also alongside them"

( Uberlieferung 257). Acceptance of this thesis by reviewers so different as Moule and Martin is significant. Substantially the same thesis was advanced by Dodd as an explanation of Johannine parallels to the synoptics (

Herrnworte 75). The same con clusion was reached by Duplacy, from his survey of recent work on the history of the NT text: "More and more, today, the redaction of the Gospels and of Acts appears as one stage, albeit essential, of written fixation in the course of a tradition partly written, partly oral, which preceded these works, contained more than they, and did not disappear after their redaction. The same tradition which moulded the sources of these works, and impregnated the minds of their authors, continued to act on the transmission of their text; the sources themselves did not disappear overnight, and they, too, can have influenced the text" ( Ou en est II. 274). See also the remarks of Robinson, ΛΟΓΟΙ ΣΟΦΩΝ 88f, on the multiplicity of sources available to the apostolic fathers.

This background, then, of oral — and in part, as Duplacy observes, written — tradition is of fundamental importance for the following study. In the first place, it prohibits

THE SECRET GOSPEL

hasty conclusions as to literary relation on the basis of occasional identities in wording.

Verbal reminiscences of the sort which in classical literature are indications of sources may here be evidence of nothing more than the common tradition of the community or the contamination of the manuscript. In the second place, as Koster pointed out in his crushing critique of Jeremias ( Herrenworte 222f), the question of priority is reduced to comparative unimportance. Once it is admitted that all Gospels alike are abstracts from the traditions of the early churches, then the fact that one was written down ten or fifteen years before or after another does not make much differ ence; the later document can easily contain the more important tradition — as Clement said the later text of Mark did. So we have two questions before us, the literary and the historical. They are at least partially independent. We now turn to the former.

Postscript , 1965: The above argument, written in 1962, can now be strengthened by the authority of Dodd and by his demonstration in Historical Tradition that the fourth

Gospel is not directly dependent on the synoptics, but derives from another, similar, body of material. Particularly important are his remarks on oral tradition (pp. 7ff) :

"It is important to realize that we are not dealing with a primitive period of oral tradition superseded at a given date by a second period of literary authorship, but that oral tradition continued to be an important factor right through the New

Testament period and beyond. Papias, in the first half of the second century, still preferred oral tradition, where it was available, and Irenaeus, towards the close of that century, could cite with great respect that which he had 'heard from a certain presbyter who had heard it from those who had seen the apostles'. . . . The early

Church was not such a bookish community as it has been represented. It did its business primarily through the medium of the living voice, in worship, teaching and missionary preaching, and out of these three forms of activity — liturgy, didache , kerygma — a tradition was built up, and this tradition lies behind all literary production of the early period, including our written gospels. The presumption, therefore, which lay behind much of the earlier criticism — that similarity of form and content between two documents points to the dependence of the later of these documents on the earlier

— no longer holds good, since there is an alternative explanation of many such simi larities, and one which corresponds to the conditions under which gospel writing began, so far as we can learn them: namely, the influence of a common tradition. To estab lish literary dependence something more is needed — some striking similarity in the use of words (especially if the words are somewhat unusual) extending over more than a phrase or two, or an unexpected and unexplained identity of sequence, or the like."

Also of great importance for what will follow here is the argument Dodd uses again and again: "It is impossible to treat any one of the Synoptics as the primary source of the Johannine version . . . since he (sic) is sometimes closer to one and some times to another of the three . . . The hypothesis of literary conflation of documentary sources seems less probable than that of variation within an oral tradition" (p. 79).

THE SECRET GOSPEL

Of greater importance for the present work than for Dodd's own are his passing remarks on the probability that elements from floating oral tradition entered the established Gospel texts as interpolations (e.g., p. 355 n2, with reference to the insertions at Lk. 6.5 and 9.56, though the great example is, of course, Jn. 7.53-8.1 1).

Further points, on particular questions, will be noted hereinafter.

The extent of agreement between Dodd's and my own estimates of the history of the composition of the Gospels is not only a source of gratification to me, but also, since the estimates were independent, a piece of historical evidence. Consequently,

I have in general left the present text just as it was before my reading of Dodd's book, changing it only by insertion of references to his work and by correction of a few errors. The reader will see that in spite of the large agreement there is consider able difference. In particular, the new material seems to me to indicate literary relation as well as dependence on similar oral traditions. How far my conclusions should be modified in the light of Dodd's study of the Johannine evidence, or Dodd's in the light of the new material, are questions for others to discuss.

II. Stylistic comparison with the canonical gospels A. Text and commentary The literary problem before us is: To determine the relation of the shorter, now canonical, text of Mk. to the longer text which Clement referred to as "the secret

Gospel" and of which Clement's church and the Carpocratians possessed somewhat different forms. This question cannot be stated as one of "authenticity." Contrast the previous question about the letter. To ask whether or not the letter is "authentic" is to ask whether or not it was written by a known individual, Clement, from whom we have a large body of original compositions which provide criteria for authenticity and make the term "authentic" meaningful when used of works attributed to him.

But "Mark," the hypothecated writer of the canonical text of the second Gospel, is not a known individual, we have no work of his save this one text, and this text is far from a wholly original composition. It combines many different kinds of material written in considerably different styles (Wohleb, Beobachtungen; Guy, Sayings; etc.).

The combination probably arose by stages (Bultmann, Geschichte, passim; Guy, Origin i22ff; Schille, Formgeschichte nf); and the present text may be thought as much the work of a "school" (a preaching or teaching tradition) as of an individual editor — Stendahl, School , has shown that even the comparatively well-ordered Mt. has such a tradition behind it. Perhaps the strongest argument for this estimate of Mk. is afforded by recent efforts to maintain the contrary: Marxsen, Evangelist , Farrer,

St. Mt ., Carrington, Mark , are principally interesting as examples of the extravagance of exegetic fantasy needed to transform " Mark" from an editor — or a series of editors

— to an author. Therefore (as indicated in the conclusion of the preceding chapter) the question before us is not to decide the "authenticity" of the new material, but

THE SECRET GOSPEL

to determine its relationship to the evidently long process of writing and editing which produced the present text of the second Gospel.

Let us begin by comparing the vocabulary and style of canonical Mk. with those of the letter's quotations from the longer text. Data for the comparison are collected in Appendix E, of which the figures are based on Moulton-Geden and Yoder and have been checked, when possible, against Morgenthaler. In the event of differences

(which were frequent) Morgenthaler's figures were preferred. It is my impression that the differences are not sufficient to affect the substance of the conclusions. I hope this will be true also of the errors which, in spite of checking, have doubtless crept in.

Collection of statistics about common phrases is difficult not only because of its monotony, which induces error, but also because of the frequency with which such phrases have been introduced into some MSS by scribal corruption. In the following pages and in Appendix E the notes on textual variants are from Nestle-Kilpatrick and (somewhat abridged) from Legg's editions of Mt. and Mk. I was not ignorant of the crticisms made of Legg's editions — e.g., Massaux, Etat 7oqf — but had nothing better to use. I was ignorant of the criticisms of the Nestle-Kilpatrick, but those I have heard to date do not seem to justify such reworking of the following figures as would be required by the change of a basic text. The data collected in Appendix E are discussed seriatim in the following commentary. They include material for com parisons with Mt. and Lk. as well as Mk. Since the style of the longer text of Mk. is very close to that of the synoptics, but has only tangential connections with that of Jn., data on Johannine usage are not included in Appendix E but are given in the following commentary when they deserve notice.

Before proceeding, one observation must be made : we know the longer text only in an eighteenthcentury copy of Clement's quotations of it. Now one of the commonest phenomena in MSS of the Gospels is harmonization of the text of one Gospel with that of the others (Williams, Alterations iff; Dodd, Historical Tradition 77 nq, 1 65E, etc.; Trocme,

Formation 169 ni); and the peculiar readings of Mk., because Mk. was the least familiar of the Gospels, suffered particularly from harmonization (cf. Hills, Caesarean

Text). Further, when Clement quoted Mk. he usually contaminated his quotations with reminiscences of Mt. and Lk. This is shown clearly by the notes in Appendix D on Clement's quotations from Mk., and even more clearly by the comparison in Appendix F between Clement's text of Mk. 10. 17-31 (III. 162. 19-163. 12) and NestleKilpatrick's. It is also interesting to see how in III.i62ff, after having quoted the Markan text correctly in his main quotation, Clement carelessly slips into the Matthaean text when he comments phrase by phrase (see especially III. 166.24). Accordingly — in spite of the κατά, λάξιν in II. 22 — we should expect Clement's quotation of the longer text to show signs of similar contamination deriving from Clement him self, and we should expect our single MS of Clement's quotations to show traces of the harmonizations of the longer text of Mk. to more familiar texts, not only of Mk., but also of the more familiar Gospels. And, last of all, it must be remembered that even the text of Clement's Stromateis rests almost entirely on one eleventh-century

MS; therefore, as Swanson observed, "it is impossible ... to determine ... to what extent the New Testament passages cited by our author have been altered, acciden tally or intentionally, by the copyists" ( Text 2).

11.23

THE SECRET GOSPEL

n.23 ^ καί βρχοντ ac els Βηθανίαν, καί ην 6Κ€Ϊ μία και έρχονται εις. Mk. has και έρχονται 4 times (2 at the beginning of a pericope—

3.31 and 12.18) and και έρχονται εις 6 or 7 times (5 or 6 times at the beginning of a pericope). Neither Mt. nor Lk. uses the expression at all. This Markan formula cannot be taken as proof of borrowing from any particular passage (cf. Turner, Usage

26.225!^) · That the subject (Jesus and the people following him) must be understood from the general pattern of Gospel stories and not from the immediate context is typical of Mk. (cf. Doudna, Greek 5ff). Another example occurs below in II 1. 6, ηλθον.

Βηϋ' ανίαν. Nesbitt, Bethany , has not succeeded in showing that the Bethany stories in the Gospels are based on a special body of traditions, but the notion is not certainly false. Mk. uses the name 4 times, Mt. 2, Lk. 2. Both of the Matthaean and one of the Lucan usages are in parallels to Mk. It would seem that the name was more prominent in the Markan material. Jn. has it 3 times in connection with Lazarus

(11.1,18; 1 2. i ; perhaps 1.28 — the text is dubious). Besides these usages, D and it. — major witnesses to the "western text" — at Mk. 8.22 have και έρχονται εις Βηθανίαν, the phrase found in the longer text. But the reading of D and it. is wrong; the place name, as given by all other witnesses, should be Beth Saida. Yet there is nothing at 8.22 to suggest Bethany. Why, then, did the scribe of the archetype of D and it. make such a blunder? Was it because he knew the longer text? D is characterized by contamination of texts from similar passages (Hatch, Text 12; Williams, Altera tions 1). και ην εκεί. Verbatim in Mk. 3.1, as the beginning of the second sentence of a story following an initial statement of place, as here. The same words, but with a different meaning ("and he stayed there") occur in Mt. 2.15. More relevant is the occurrence in all three synoptics of ην/ησαν δε εκεί as a narrative formula: Mt. 27.55,61; Mk.

5.1 1 I] Lk. 8.32. Matthew has introduced the formula where Mk. did not have it.

It is a formula, not a sign of literary dependence. [W.M.G. questions this conclusion, remarking that Mk. 3.1 also has the same sentence structure as this phrase of the longer text: copula, verb, adverb, subject, attribute.] But substantially the same sentence structure is also found in Mt. 27.55,61 and Lk. 8.32. Evidently it went with the formula. μία. Mk. has this adjectival use of εις for τις in 12.42, ελθονσα μία χήρα πτωχή, but it is more common in Mt. (4 instances, with a fifth in the western text, Ode

Sy.c,s* Arm.). Hawkins ( Horae 4, 30Q lists it as a Matthaean trait, but it is probably a Semitism (no instances in Lk.), not a sign of literary dependence. (There are occasional occurrences in classical Greek — cf. Aristophanes, Aves 1292, and van

Leeuwen's note — but the Semitic usage was so common that its influence is probably to be supposed.) εις for τις in other constructions ( εις των προφητών, εις εκ του δχλον, etc.) is common in all the Gospels, as in the papyri. Since the decision as to the number of instances depends on the subjective judgment, whether or not the author

11.23

THE SECRET GOSPEL

\ T f 5 O \ / ' ~ γυνή ης o a θ€λφος αυτής intended to emphasize the singleness of the object referred to, there will be differences of opinion about the lists in Appendix E, which give 14 or 15 instances in Mk.,

14 or 18 in Mt. and 13 or 15 in Lk. Cf. Moule, Idiom-Book 125 and 176; BlassDebrunner-Funk, no. 247.2; Doudna, Greek 33b γυνή. Women come to Jesus in Mk. 5.27 (the woman with an issue); 7.25 (the Syrophoenician) ; 14.3 (the anointing in Bethany); 16.2 (the resurrection). To these

Mt. adds 20.20 (the mother of James and John) ; Lk. adds 7.37k (the Lucan anointing

—that the woman came is implied if not stated); 10.40 (Martha's complaint); 23.27 (the women of Jerusalem) ; Jn.: 11.20 (Martha, in the raising of Lazarus);

11.32 (Mary, in the same); 20.1 (the Johannine resurrection). Besides these the adulteress is brought to Jesus (Jn. 8.3) and there are a number of meetings with women or mentions of them in Jesus' entourage. Evidently they played a large part in early

Christian tradition. The multiplicity of these stories makes it impossible to be sure that the story in the longer text of Mk. was drawn from any one of them. ης . . . αύτης. Redundant αυτός following ός in the oblique cases is found twice in Mk., once in Mt., and twice in Lk. (one Markan), always in the genitive, ης . . . αύτης appears only in Mk. 7.25. The same construction appears again in III. 15, below, in the accusative. It is probably a Semitism rather than a sign of literary dependence; there are 10 instances, in all three oblique cases, in Apoc. (These figures do not include the peculiar readings of codex Bezae; Yoder's concordance has not indicated the peculiar usages of αυτός.) Both the instances in the longer text, and all those in canonical Mk., have in common a trait which Doudna was not able to find in the papyri, "namely, the fact that the redundant possessive pronoun follows its noun immediately" [Greek, 38). See also the note on III. 15, ον ηγάττα αυτόν. γυνή ης . . . αύτης . . . ελθοΰσα. This verbal sequence appears also in Mk. 7.25 (the Syrophoenician). However, it would be hasty to conclude that either passage is dependent on the other, because: (1) γυνή . . . e'A θοϋσα appears again in Mk. 5.25k;

(2) ελθοΰσα in our text is not part of the same sentence as γυνή ης . . . αύτης but is part of the formula ελθοΰσα ττροσεκύνησεν, of which there are 7 variants in Mt. (see below); (3) αύτης in Mk. 7.25 is omitted by RDW/ΙΘ Pap. 45, fam.\,fam. 13(^.124),

28.225.237.253.475**.565.569.7oo.tf/.jfrtfW£.zhvg.Cop.sa'bCKGeo. (so Legg). Now if we suppose the longer text of Mk. to have been compiled from texts of the types known to us, we should suppose the compiler to have used a text akin to the archetype of D and it., since only from such a text could he have gotten και έρχονται εις βηθανίαν

(above). But the archetype of D and it. evidently did not have the redundant αύτης in Mk. 7.25. So there is some difficulty in supposing that both similarities resulted from imitation.

II.23-24

THE SECRET GOSPEL

II.24 άπβθανβν και βλθοΰσα προσβκυνησβ άττεθ ανεν. The form: 2 in Mt., 6 in Mk., 5 in Lk. This perfective meaning for the aorist ("had died" or "was dead") appears in Lk. 8.53 and possibly in Mk. 9.26 and 15.44 (in spite of the fine distinction made by Swete and accepted by Taylor; the emendations — τεθνηκεν W@ 472.1342 and τεθνήκε i D,/253 — show how the copyists understood the Greek). It is regular in LXX: Ex. 16.3; Num. 14.2; 20.3; Judges

8.33b; 9-55 (?) ? II Sam. 13.39; etc· The sentence structure here, with the verb at the end, is thought typically Markan by Turner, Usage 29·352ίΓ. καί + participle of έρχομαι in the nominative + finite verb: 5 in Mk., 12 in Mt.

(only one Markan) and 5 in Lk. (none Markan). These figures would be larger if account were taken of the instances with δε instead of καί or with compounds of έρχομαι (εις-, προσ-, εξ-) . This standard construction cannot be taken as evidence of dependence on any particular passage. For the same construction with other verbs see below, on καί όργισθείς (II. 25). The use of redundant participles, especially ελθών and άναστάς (III. 10), is studied by Doudna (

Greek , 55ff and 1 1 7fT) as charac teristic of Mk.'s style. προσεκυνησε και λεγει. [W.M.C. notes the connection — by καί — of an aorist with a historical present and remarks that, though unusual, it does appear in Thuc.,

VI. 4. i.] In Mk. it is common, particularly so with λεγει (1.37,41,43!, etc.), as here in the longer text. In III. 7, below, it recurs with ερχεται, as in Mk. 6.1. [For a study of earlier examples W.M.C. refers to von Fritz, Present esp. 195b] προσεκυνησε. προσκυνεΐν after έρχομαι'. 7 or 8 in Mt., never in Mk. or Lk. In 5 of the 7 instances it follows a participle. One instance (15.25) is particularly close to the above: η δε ελθοΰσ a προσεκυνει a ύτω ( ττροσεκννησε LKXd Φ 0119V Ι57·5^5·β^· pier. it. pc. wg. pier. Sy.Cop.b0,Aeth.Or. ; αυτόν /1174.1515). This again is the story of the Syrophoenician woman. If the longer text of Mk. was compiled from the canonical texts, the compiler took his first sentence from Mk. 7.25 and his second from Mt.

,15.25, which he revised by substitution of initial καί for δε— not a likely procedure.

Matthew's interest in proskynesis is remarkable; but the verb cannot be taken as proof of dependence on Mt., especially since it here governs the accusative, which it never does in Mt. except when he is quoting the OT. Mark also represents suppliants as approaching Jesus with proskynesis (5.6, governing the accusative) and might have used the καί ελθών formula to introduce them [R.S. calls attention to Mt.

20.20, τότε προσηλθεν a ύτω η μητηρ των υιών Ζεβεδαίου μετά των υιών αυτής προσκυνούσα καί αίτοϋσά τι. This introduces a proskynesis by a woman into the story (= Mk.

1 0.35—45) which, according to Clement, immediately followed the present quotation from the longer text. Moreover, the woman introduced — the mother of the sons of Zebedee — replaces Salome in Mt.'s parallel to Mk.'s list of the women who witnessed the crucifixion (Mt. 27.56 || Mk. 15.40). Her request, in Mt. 20.20, was

11.24

THE SECRET GOSPEL

τον Ίησονν refused. Salome is one of the women who appears in the next section of the longer text (immediately following the pericope introduced by Mt. 20.20) ; there she comes to see Jesus and is refused.] This suggests that Matthew incorporated in 20.20 traces of the preceding and following stories of the longer text of Mk. His introduction of the mother of the sons of Zebedee is commonly explained as a consequence of reverence for the apostles : where Mk. reported their ambition and rebuke, Matthew reported their mother's ambition on their behalf (so, for example, McNeile). But Matthew associated the sons with their mother's request, and the rebuke is still addressed to them (Mt. 20.22ff). And it is Matthew's way, when abbreviating Mk., to use the introduction of an omitted story at the beginning of the next story; for example, he retains Mk. 5.18a (= Mt. 9.1a), omits 5.i8b-20, and uses 5.18a to introduce Mk. 2.1 ( = Mt. 9. ib). He also takes details from omitted episodes and uses them elsewhere: thus Mk. 1. 21-22 in Mt. 7.28b(f); Mk. 1.24, ήλθες, in Mt. 8.29;

Mk. 1.28 in Mt. 4.24a; etc. Therefore it is noteworthy that, besides introducing in 20.20 his substitute for Salome— making proskynesis — he has also introduced into the story of the Syrophoenician woman the appeal not found there in Mk. : ελεησόν με, κύριε, υιός Δανειδ (Mt. 15.22). The story of the Syrophoenician was that of which the beginning most closely resembled that of the story in the longer text where the appeal is found. These two pieces of evidence (20.20 and 15.22) fit together and suggest that Matthew knew the longer text of Mk. Somewhat similar conclusions have been reached on other grounds by several scholars (Parker, Gospel·, Vaganay,

Absence', etc.). Further evidence suggesting that Matthew may have known the longer text of Mk. appears below, on άπεκύλισε (III. if). τον Ίησοΰν. The accusative after προσκυνεΐν was the classical Greek construction

(Blass-Debrunner-Funk, no. 15 1.2) and appears in Mk. 5.6, where, however, there is minority support for the dative. In Mt. and Lk. it is found only in quotation of Dt. 6.13 (Mt. 4.10 = Lk. 4.8, κύριον τον θεόν σου προσκυνήσεις) . However, besides the variants noted in Appendix E, most relevant passages have one or two minuscule variants replacing the dative by the accusative; this was the tendency of later Greek.

In studying the letter we saw that its chief difference from Clement's usage lay in the preponderance of accusatives after prepositions. If some accusatives there and the one here were introduced by a medieval copyist, they would hardly have been eliminated by a later corrector. [T.B. remarks that "in Aik. 5.6 it is especially the Alexandrian group that testifies for the accusative: BCL ΔΨ 892,1241. Beside these

MSS we find additional testimony in the I-group of von Soden: A, 047, 1 79,230,273,

482,495,544,659,700,1346,1574,1588,1606, of which A, at least, testifies to the Alexandrian reading. This would be an indication for the Alexandrian source of the secret gospel."] Or of classicism in the Alexandrian revision? But see below, on και ευθύς (III. i ) , where the agreement with the Alexandrians is in a detail contrary to classical usage.

Π.24-25

THE SECRET GOSPEL

Π.25

Δαβίδ βλέησόν μβ. oi Se μαθηταί /cat λβγβι αύτώ vie καί Χεγει αύτώ. 8 in Mk. ; in Mt. 6 or 7 (only 2 Markan) ; never in Lk. Not evidence of dependence on any specific passage, λέγω is often used in the Synoptics to introduce e\ εησον (but almost always with some other verb of utterance, usually κράζειν — a fact to which my attention was called by T.B.). In Mt. 17.15, however, it stands alone, as here. The parallelism probably results, not from dependence, but from the presence in both instances of another verb form (indicating motion). Three verbs together are cumbersome.

VL€ . με. This form of appeal is found in Mk. 10.47,48 || Lk. 18.38,39. In Mt.

9.27; 15.22; 20.30,31 (although 20.30,31 are Markan and 9.27 may be) the wording has been changed, probably under the influence of the liturgical use of κύριε εΧ εησον

(also used in appeals to pagan deities, Epictetus, Dissertationes II.7.12). On 15.22 see above, s.v. προσεκύνησε end. Clement, II.498.32ff, arguing that gnosis is not to be made available to everyone, says, άμεΧει και τών επιβοωμενων τον κύριον αυτόν οι μεν ττοΛΛοι vie Δα pio, ελεησον με ελεγον , ολίγοι οε υιόν εγιγνωσκον του σεου , κασαπερ

6 Πέτρος , ον και εμα κάρισεν. This is surprising because the canonical Gospels contain many instances of Jesus' being recognized as son of God (Mk. 3.1 1 ; 5.7 and parallels;

15.39 and parallels; Lk. 4.41 ; Mt. 14.33; etc.), while the appeal to the son of David appears only in Mk. 10. 47k and parallels, Mt. 9.27 and 15.22. Perhaps Clement may have had in mind also Mt. 21.9 and 15 — see his vague exegesis in I.97.10-13.

Nevertheless, his mistake is surprising in one who knew the Gospels so well. Surprising, too, is his choice of the Markan form of the appeal rather than the more frequent form in Mt., his favorite Gospel. Are both these surprising details to be explained by his knowledge of additional Markan material, including the selection quoted in the letter, in which other speakers beside Bartimaeus appealed to Jesus as "son of David"? [W.M.C.: The omission of ώ before the vocative would, in classical

Greek, have been impolite and indeed insulting.] In the Gospels ώ with the vocative has become a sign of emotion (Blass-Debrunner-Funk, no. 146). In Mk. it appears only in ώ γενεά άπιστος , None of the eight canonical appeals to the son of David (listed above) has ώ. οι δε μαθηταί. Initial, with an immediately following verb, in Mk. 10.13 and 24.

Mark has a habit of using the same construction several times in quick succession : see the distribution of ευθύς in his text. If the quotations from the longer text stood where the letter says they did, their usage of ol δε μαθηταί would form a third member of the above group. Mt. 19.13 is Markan, but Mt. has the construction independently in 1 2. i (whence it has been taken into Mk. 2.23 by a few MSS) and in 28. 16 ( -f ενδεκα)

and, modified, in 14.26. D<9 it. have it in Mk. 14.4, where it may be original — censor ship (altering passages discreditable to the Holy Apostles) would explain the common text. Never in Lk. ; in Mt. and Mk. evidently a standard locution.

11.25

THE SECRET GOSPEL

επβτίμησαν a vrfjm καί οργ ισθείς επετίμησαν. The verb, 9 times in Mk., 6 or 7 in Mt. (all Markan), and 12 in Lk.

(5 non-Markan). In Mk. and Mt. rebukes are connected with the danger of revealing secrets. The demons are rebuked that they should not make him known, Mk. 3.12 and parallels; the disciples are rebuked that they should not declare what Peter has said, Mk. 8.30 and parallels; Peter rebukes Jesus for teaching openly that the Son of Man must die, Mk. 8.32 and parallels (note the reading in c and k, ne cui haecjilla diceret)', Ebeling, Messiasgeheimnis 1 36Γ, thinks the followers of Jesus rebuked Bartimaeus for declaring Jesus the son of David, Mk. 10.48 and parallels. That the disciples should have rebuked Lazarus' sister for blurting out the same title is therefore in accordance with Markan practice. ol δε μαθηταί επετίμησαν. Verbatim in Mt. 19.13 and many MSS of Mk. 10.13 — combination of a common verb with a standard locution for a customary purpose. καί + nominative participle + finite verb. One of the most common sentence structures in the Gospels. Matthew was fond of it with participles of έρχομαι, (see above, on καί ελθοΰσα, II. 24), but a check of occurrences with all verbs — in Mk.

1-3 and 10, Mt. 4, 8, 9, 19, and 20, and Lk. 4, 5, 6.1-11, and 18 — suggests that it is most common in Mk. In these chapters Mk. has 33 instances, against 22 in Mt.

(10 Markan) and 20 in Lk. (8 Markan). καί, participle, ό ' Ιησούς, verb, without interruption, occurs in these chapters 3 or 4 times in Mk., 2 in Mt., and 2 in Lk.; but both Lucan instances are the cliche καί άποκριθείς 6 Ίησοΰς είπεν. όργισθε ίς. [Mark did not eliminate emotions of Jesus, as the other evangelists did: cf. Mk. 1.41,43; 7.34; 8.12; 10.14,16,21 with the parallels. R.S.] Hawkins, Horae

1 19, notes that except for the western reading in Mk. 1.4 1, οργή is nowhere in the Gospels ascribed to Jesus save in Mk. 3.5. Therefore, although Mt. uses the verb οργίζω 3 times and Lk. 2, their usage is probably irrelevant to the use here in connec tion with Jesus. The aorist passive participle occurs in Mt. 18.34 ( καί οργ ισθείς 6 κύριος αύτοΰ παρεδωκε ν) and in Lk. 14.21 in a similar construction: τότε όργισθε ίς ό οικοδεσπότης είπεν. It is used of Jesus in the D text of Mk. 1 .4 1 : καί όργισθείς εκτείνας την χεΐρα αύτοΰ ηφατο αύτοΰ, with support from it., Tatian, and Ephraem. However, όργισθείς in Mk. 1.41 is hard to explain (Eitrem, Demonology 4.2) , whereas σπλαγχν ισθείς, the reading of all other MSS, fits the context (which contains no trace of opposition).

In the letter's longer text of Mk., however, όργισθείς is easily explicable: Jesus could have been angered either by the use of his secret title or by the disciples' rebuke of the suppliant (cf. Mk. I0.l3f). καί προσεφερον αύτω παιδία ΐνα αύτών αφηται' οι δε μαθηταί επετίμησαν αύτοΐς. ιδών δε 6 Ίησοΰς ηγανάκτησεν. If the Story in the longer

text is of the raising of Lazarus, its όργισθείς corresponds to ενεβριμησατο in Jn.

11.33. And εμβριμησάμενος occurs in Mk. 1.43. Perhaps the archetype of DzY. intro duced όργισθείς into Mk. i. 4 1 inadvertently, by reminiscence of the longer text

THE SECRET GOSPEL

II.25-26 II.26

6 'Ιησούς άπηλθεν μετ' αυτής εις τον κήπον οπον ην το μνημεΐον. and anticipation of εμβριμησάμενος. The reminiscence would be facilitated by the fact that the following words in Mk. — εκτε ίνας την χ€Ϊρα αυτοϋ ηφα το αύτοϋ — are soon paralleled in the longer text: εζετεινεν την χεΓρα καί ηγειρεν αυτόν. Βηθ ανίαν in II. 23 suggested that the archetype of D it. in Mk. 8.22 was corrupted by a reminiscnece of the longer text. Since DzL are representatives of the western text, which was used by Clement, and since Clement knew the longer text, these suppositions are not unlikely. It is interesting that Westcott and Hort, II. appendix. 23, supposed όργισθείς in Mk. i. 41 "perhaps suggested by verse 43, perhaps derived from an extraneous source." Can it be that we now have the source? [C.F.D.M. thinks more likely a suggestion he heard from W. Howard: that οργισθείς in Mk. 1.41 originated as a marginal gloss on εμβριμησάμενος.] άπηλθεν. This form, 9 in Mk., 7 in Mt. (2 Markan), and 6 in Lk. (2 Markan).

It is followed by μετά with the genitive only in Mk. 5.24. There, as here, Jesus goes off with a suppliant (Jai'rus) to raise a dead relative. However, the construction is normal and can hardly be taken as evidence of literary dependence. LXX: II Sam.

16.17; To. 14.12; Siracides 14.19. Άπελθεΐν els is common. κήπος. In the synoptics only in Lk. 13.19, όμοια εστίν ζη βασίλεια τον θεού} κόκκω σινάπεως , ον λαβών άνθρωπος εβαλεν εις κήπον εαυτοΰ. This may be an adaptation of the mustard-seed parable to the story of Jesus' burial in and resurrection from a κήπος, Jn. 1 9.4 iff. ην δε εν τω τό πω όπου εσταυρώθη κήπος, καί εν τω κη πω μνημεΐον καινόν, κ.τ.λ. Interpretation of the Lazarus story as foreshadowing Jesus' resurrection may have led to the location of Lazarus' tomb, too, in a κήπος. Tombs in gardens near Jerusalem are mentioned in Josephus, AJ IX. 227 and X.46. [Burial in gardens was common in the Greco-Roman world and was reflected by the Greek words, κηποτάφιον and κηπόταφος or κηπόταφον, and the Latin cepotaphium, LSJ s.vv. A.D.N.] όπου ην. όπου, 1 5 times in Mk., 13 in Mt., 5 in Lk. όπου rjv occurs in Mk. 5.40 (as here, of Jesus' going to raise the dead), in Mk. 2.4, and twice in the D text of Lk.

(4.16 and 5.19 — the second Markan). 6 uses in Jn. One is in the Lazarus story (1 1.32), one in a reference to Bethany όπου ην Λάζαρος (i2.i), and one in a reference to the κήπος where Jesus was arrested (18. 1). It is difficult to be confident that the expression in the longer text of Mk. was taken from any one of these possibly relevant passages. το μνη μεΐον . 6 in ML., 7 in ML, and 7 in Lk. (MLrgenthaler) , Yoder adds 3 in

Mk. and 3 in Lk., from D. The only reference to Jesus' having raised a dead man from a tomb is in the Lazarus story, where μνημεΐον occurs 3 times (Jn. 1 i.i7,3L3^) and is subsequently mentioned in the popular report of the miracle (12.17)·

II. 26-111. i

THE SECRET GOSPEL

III. i /cat ευθύς ηκούσθη εκ τού μνημείου καί εύθύς. As the beginning of a sentence or independent clause, 25 in Mk., 2 in Mt. (both instances Markan), never in the best supported text of Lk. Of the instances in Mk., D has only 1.30, 4.5, and 11.3. D omits both the well supported instances in Mt., but is unique (?) in reading καί εύθύς in Mt. 13.5 (also Markan) and almost unique in reading it in Lk. 5.6. Otherwise it consistently substitutes ευθέως, or omits.

The longer text's repeated use of εύθύς, therefore, links it not only with Mk., but with a small MS tradition of the Markan text — once again that of the Alexandrian

MSS. See Appendix E, and T.B.'s remark on τον Ίησοΰν above, on II. 24. Since the longer text is right, against D, in this characteristic, it seems more likely that the corruptions of the D text noted above (Βηθ ανίαν and οργισθείς , cf. 1 1.2 3 and 25) were derived from the influence of the longer text, than that the latter derived these details (which, in it, do not seem to be corruptions) from the archetype of D, to which they were peculiar, καί εύθύς immediately followed by a finite verb, 1 in Mt., 9 in Mk. The entry in Moulton-Geden for Mk. 7.35 is contradicted by the readings in Legg. In the present usage in the longer text, ενθυς seems to be a connective rather than an adverb of time; so it is in Mk. (Kilpatrick, Notes 4Q. ηκούσθη. The form occurs in Mk. 2.1 and in Mt. 2.18, where the subject is, as here φωνή — ( φωνή εν 'Ραμα ηκονσθη, LXX Jer. 38· 1 5) — probably coincidental. [T.B remarks that in some "western" texts ηκονσθη is added where the normal Greek texts read only φωνή: so Mt. 3.17 (Sy.s·) ; 17.5 (Sy.c·) ; Mk. 1.11 (0.pauc.); Lk. 3.21

(Sy.s').] Influence? άκονω follows ευθύς in Mk. 7*25, ύλλ' ευθύς άκονσασα γυνή περί αυτού (XBLd 33·579·^925 ευθέως D it. ; άκονσασα γάρ rell. ), in the story of the Syrophoenician woman, which paralleled the letter's Gospel above in II. 23-24. However, given the difference of the constructions and the frequency of both ευθύς and ακούω in Mk., this similarity, too, is probably coincidental. καί ευθύς ηκούσθη. The following story shows many traits in common with resur rection stories — voice, youth, stone, etc. On these see the comment below on ό νεανίσκος, III. 3. εκ τού μνημείου. The local sense of εκ after ακούω does not occur in the Gospels.

(Cf., however, Jn. 12.34.) But the construction is common: Apoc. 10.4.8; 11.12; etc. On μ νημεΐον see above, on II. 26. εκ τού μνημείου does not appear in the synoptics

(though Mt. and Mk. have εκ των μνημείων) , but is used 3 times in Jn. Note Jn.

12.17, τον Λάζαρον εφώνησεν εκ τού μνημείου, and 3 · 2Bf, πάντες οί εν τοΐς μνημείοις άκούσουσ ιν της φωνής αύτού καί εκπορεύσονται. These verses reflect the Johannine version of the story, in which Jesus cries out; in the longer text of Mk. the voice comes from the tomb. This contentual relationship will be discussed later.

THE SECRET GOSPEL

III. 1-2 III.2 φωνή μεγάλη, καί προσελθών 6 ' Ιησούς άπεκυλισε τον λίθον φωνή μεγάλη. \ in Mk., 2 or 3 in Mt., 6 in Lk. (one is omitted by D, but another is added). Of the 3 Matthaean usages, the 2 certain ones are Markan; of the 5 certain Lucan usages, 3 are Markan. All the uses in the synoptics are in oblique cases. The phrase is used in Jn. only once, in 11.43: φωνή μεγάλη εκρανγασεν <ζό

Ιησούς), Λάζαρε , όεΰρο εξω. Since the phrase was not a common element in John's vocabulary but was in Mark's, John is more likely to have derived it from a Markan story of the raising of Lazarus than Mark from a Johannine one. (The rarity of the phrase in Jn. is the more remarkable because of his fondness for φωνή in other constructions, noted by Dodd, Historical Tradition 282.) καί προσελθών. Initial, 1 or perhaps 2 in Mk., in Mt. 5 or 6 + 1 in D, in Lk. 2.

Of these instances the dubious one in Mk. corresponds to the dubious one in Mt. ; the rest are independent — evidently the locution was standard. In Lk. 7.14 (the young man of Nain) the actor is Jesus and the following action is a raising from the dead, as here; but there is no other contentual similarity, so these are probably irrelevant. In Mk. 1.31 the following words are ηγει,ρεν αυτήν (Peter's wife's mother), but she was raised only from a sickbed. By requiring exact parallel in form and position, the above figures obscure Matthew's characteristic preference for προσέρ χομαι. Moulton-Geden list 52 uses in Mt., 6 in Mk., and 11 in Lk. ; Morgenthaler lists 52 uses, 5 and 10, respectively. καί προσελθών 6 Τησοΰς. Verbatim, as the beginning of a sentence, in Mt. 28.18.

Followed by the charge to make converts of all nations. Here also we have καί προσελθών + explicit subject + verb, without interruption, as in Mt. 4.3 and 8.19 and the longer text of Mk. καί προσελθών άι τεκνλισε τον λίθον. Found in the middle of a verse, in Mt. 28.2

(the angel at the resurrection), καί is omitted by many MSS, including D. One form or another of άποκυλίειν + τον λίθον appears once or twice in both Mk. and Lk., the usages in Lk. being Markan. The λίθος closing the entrance of the μνημεΐον appears also in Jn.'s raising of Lazarus (ii.38ff) and there, too, is removed. Mt.

28.2 is not a parallel to Mk., but may be an attempt to explain what Mk. reports.

Mt.'s angel of the Lord coming to raise Jesus is much like Jesus coming to raise

Lazarus. Where did Matthew get the idea? Not from the Markan resurrection of Jesus; nor from the Johannine story of Lazarus. Perhaps from the longer text of Mk. ? That he did not copy thence από της θυρας τον μνημείου is not surprising.

He did not copy it from Mk. 16.3, either; his way was to abbreviate. The details of

the Style in the longer text suggest Mk., άπεκυλισε . . . από , μνημεΐον . . . μνημεΐον . . . μνημεΐον. Compare the examples at the end of the note on κρατησας της χ^ιρός, below, on I II. 3-4.

111.2-3

THE SECRET GOSPEL

OL7TO της θύρας του μνημείου καί είσελθών εύθυς οπού ην από της θύρας του μνημείου. Verbatim, Mk. 16.3 CDW01/7 fam. 1 3 (exc. 1 24) *543· 1 57*

(cor.**). 14.472.51 7. Eus.dem,Greg.Nyss., with ab, it.vg.; εκ XABLX/M/7T1? minusc. pier. The reading από , native to Mk., crept by contamination into Mt. 28.2. The contaminated reading is exactly that of the longer text with από, not εκ. The stages of the contamination are shown in Appendix E. When complete it produced καί προσελθών άπεκύλισε τον λίθον από της θύρας του μνημείου, word for word parallel to the longer text, except for the absence of ό * Ιησούς . But this complete form appears only in the later versions and uncials, and in minuscules. Therefore — unless it be one of the rare examples of early readings found only in late MSS (Williams, Altera tions 32) — it cannot have influenced the longer text, already known to Clement, nor can it reflect the influence of the longer text, almost unknown after Clement's time. Thus one of the most striking parallels between the longer text and a canonical

Gospel cannot be explained by direct dependence on either side. Among the many possible explanations are (1) that the author of canonical Mk. also wrote the longer text and repeated himself — as he often did in canonical Mk.; (2) that the longer text was produced by the same process of conflation which later produced the contami nated text of Mt. ; (3) that the longer text was conformed to that of Mt. by some medieval copyist. [H.K. remarks that Lk. 24.2 also has από του μνημείου.] Therefore, it is likely that Mt. and Lk. reflect a text of Mk. which had από rather than εκ, but the phrase is so commonplace that no secure conclusion is possible. καί είσελθών /οΰσα (etc.). Initial, 5 or 6 in Mk. + 1 in D and Sy.s·; 1 in Mt. ; 4 in Lk. (1 Markan) 41 in D. A standard introductory phrase most favored by Mark; hardly useful as a sign of authorship and not evidence of borrowing from any par ticular passage. This conclusion is confirmed by the frequency of εισέρχομαι in all forms: Mt. 36, Mk. 30, Lk. 50 (Morgenthaler). καί είσελθοΰσα ευθύς, initial, appears in Mk. 6.25, but the passage (Salome's approach to Herod) has no contentual similarity to the text here. Therefore, given the frequency of both εισέρχομαι and ευθύς, the coincidence is probably accidental, είσελθών, without καί and not initial, introduces in Mt. 9.18 the raising of Jai'rus' daughter — which has a number of verbal parallels to the present passage, probably because of similarity of content.

See below, on κρατησας. όπου rjv. See above, on II. 26. Mk. 2.4 and 5.40. The second occurs in the raising of Jai'rus' daughter, where it follows είσπορεύεται and precedes, as here, κρατησας της χειρός. The conjunction of these conventional phrases in the same sequence is explained by the common content of the passages. [W.M.C. remarks that in είσελθών

. . . όπου the εις is pleonastic, ελθών would suffice.] However, as already stated, Mk.

5.40 has the same construction with είσπορεύεται; and similar pleonastic constructions with είσπορεύομαι and εισέρχομαι preceded by οπού occur in Mk. 6.10,56, and 14.14 but not in the other Gospels. This seems to be a Markan trait.

III.3-4 III.4

THE SECRET GOSPEL

6 νεανίσκος. 2 in Mk., 2 in Mt., i in Lk. All of these νεανίσκοι, have traits which relate them to the present story. The one in Mk. 14.51, wearing a sheet over his naked body, was (almost) caught with Jesus late at night; the one in Mk. 16.5, wearing a white garment, was found in Jesus' tomb and announced his resurrection; the one in Mt. ig.2off was loved by Jesus (Mk. 10.21, εμβλεφας αύτω ηγάπησεν αυτόν ) and was rich (Lk. 18.23, Vv Ύ^Ρ πλούσιος) ; the one in Lk. 7.14 was raised from the dead, and the story contains the verbal sequence καί προσελθών ηφατο . . . νεανίσκε . . . εγερθητι (but this is probably mere chance). Another νεανίσκος was raised or saved from death in the D text of Acts 20.12. Yet none appears in a story so close to the present one as the story of Lazarus, and in the canonical Gospels

Lazarus is not called a νεανίσκος. This multiplicity of partial parallels suggests narration in traditional patterns with traditional vocabulary, rather than compilation from written sources. [H.K. agrees, remarking that a similar variety of parallels — voice, youth, stone, etc. — appears also in the earliest apocryphal Gospels, especially

Peter, and probably for this same reason. For a study of these parallels he refers to L. Brun, Die Auferstehung Christi (1925), which I have not seen.] εζετεινεν την χεΐρα. εκτείνειν + χειρ, 3 or 4 in Mk. (one with χεΐρα merely implicit),

6 in Mt. (3 Markan, χεΐρα once implicit), and 5 in Lk. (2 Markan, 2 from D).

The exact combination εξετεινεν την χεΐρα seems to have stood in the achetype of Sy. and some MSS of it. at Mt. 12.13. The story there (the man with the withered arm) has no similarity to the present one, so the identity of wording was probably coincidental. More striking are the western variants to Mk. 1.3 1, the only place where εκτείνω precedes εγείρω. On these see below, on κρατησας. Taking the hand and raising the dead is frequent in Acta Ioannis : 1 1, 47, 79, 83, 1 12. There are many pagan parallels (emperors raising the afflicted, gods welcoming the deified dead, etc.); see Schrade, Ikonographie iogff. ηγειρεν. εγείρω Mk. 19; Mt. 36; Lk. 1 8. ηγειρεν appears only in an OT reminiscence in Lk. 1.69 and in Mk. 1.31 and 9.27 (ηγειρεν αυτόν) ; on these see the following note. καί ττροσελθόντες ήγειραν αυτόν in Mt. 8.25 (stilling the storm) is a coincidence, as is κρατήσει αυτό καί εγερεΐ in Mt. 1 2. 1 1 (the man with the withered arm) and νεανίσκε , σοι λέγω, εγερθητι in Lk. 7.14 (the young man of Nain). κρατησας της χειρός. Verbatim, 2 or 3 in Mk., I in Lk. (Markan). κρατεω, Mk.

15, Mt. 12, Lk. 2. χειρ as an instrument of supernatural help, 10 or 11 in Mk. ( + 1 in the long ending), 7 in Mt. (5 Markan), 5 in Lk. (2 Markan + 1 in DzT). ά-πτω in the same use, Mk. 1 1, Mt. 9, Lk. 10. Neither χειρ nor άτττω ever has this use in lop Jn. The exact phrase κρατησας της χειρός is always used in the Gospels in connection with εγείρω :

111.3-4

THE SECRET GOSPEL

Mk. 1.31 ( Peter's mother-in-law) || Mt. 8.13; Lk. not similar.

W καί ττροσβλθών £κτ ίνας (sic) τήν χεϊρα καί όπιλαβόμ€νος ήγειρεν αυτήν, bq ille autem venit et extendens manum adprehendit eam et levavit, r 1 et veniens extensa manu adpraehensam elevavit eam.

Mk. 5.40/ (Janus' daughter ) || Mt. g.25 || Lk. 8.34.

KB, etc. καί εισπορευεται όπου ήν το ττα ιδίον καί κρατήσας τής χειρός του παιδιού Aeyei . . . eyeipe.

DzT καί eianopeoeTo όπου ήν το παιδιού καί κρατήσας τήν χεϊρα του παιδιού (eius, it.) Aeyei . . . eyeipe.

D όλθων €κράτησ€ν τήν χεϊρα αυτής καί ήγόρθη τδ κοράσιον. it.vg. intravit et tenuit manum eius et dixit puella surge.

Mk. g.2y (the demoniac boy). Mt. and Lk. not similar.

KBDLZl ΘΨ f am. i. Jam. 13 (exc. 1 24) .28.53.543.565.892.7.^g.Sy.hier· Cop.sab0Geor.

ΑΟΝΧΥΓΠΣΦ^ 22.ι 24.33· I57-579-700-I07I-^*/,^r*Sy.s'pesh'hl' Aeth. δ Se 7 ησοϋς κρατήσας αυτόν τής χειρός (+ αύτοΰ C*Sy.s etc Aeth.) ήγειρεν αυτόν.

(On Mk. i. 3 1 see Couchoud, L'Evangile 174. Pernot, Pretendu 50, supposes D influenced by Mk. 1.4 1, but this is unlikely: the verses are not sufficiently similar. Couchoud's reply, Marc latin 294.) Here the words in boldface type appear in the same forms in the longer text of Mk. To attribute this fact to deliberate compilation from the multiplicity of written texts is implausible. Evidently the similarity of the longer to the shorter text of Mk., like the similarity of the different passages of the shorter text to one another, is the product of free composition in a standard form and with a standard vocabulary.

It is hard to believe that any compiler would have produced the awkward repetition χεϊρα . . . χ€ΐρός. The D it. text of Mk. 1.31 might be a deliberate revision of this, and the W text a further revision of D. Can it be that here, again, memory of the longer text contaminated the archetype of D and produced a text reading e/creiVa? τήν χεϊρα κρατήσας τής yei ρός ^yeipev αυτήν, of which various corrections now appear in DWzi? Pleonasm is a well known trait of Markan style (Hawkins, Horae 1 39fT) ; so is verbal repetition (above, on μνημ€Ϊ ον, in III. 2). Both the later synoptists and the later MS tradition tended to eliminate these traits by choosing one or the other half of Mark's redundant expressions. For example:

1.26

I-34 €κ . . . e£eA θόντ€ς, om. Mt. Lk. ; e\ θόντ€ς 73· δαιμόνια . . . δαιμόνια, om. second δαιμόνια Lk.0Dz7.yg.Sy.s'Aeth.

1 10

III.4

THE SECRET GOSPEL

6 δε νεανίσκος εμβλεφας αύτώ ηγάπησεν αυτόν ΐ·35 εξήλθεν καί άπηλθεν, om. άπηλθεν Lk.B28*<2/. ; om. εξήλθεν καί W it.vg.

1.43 α ύτώ . . . αυτόν, om. Mt., Lk., W b.c; om. αύτώ 349.517 a.e.vg.; οπι. αυτόν

255Sy.s· 2.9-12 άρον/άρας τον κράββατον ter, om. first Mt., Lk., W 544. 692.fi. ; om. third

Mt.

2.15—16 τελώναι καί αμαρτωλοί ter, om. second Mt., Lk., W; τελωνών tant. 6gal.

2.18—19 νηστεύοντες . . . νηστεύουσιν . . . νηστεύουσιν . . . νηστεύειν . . .νηστεύειν, om. νηστεύοντες Mt., Lk. ; om. second νηστεύουσιν Lk. 543 it. pane.', om. first νηστευειν Mt.; om. second νηστευειν Mt., Lk. DU^ W fam. 1 [exc.

I3I)33-700.^-^r*Sy.pesh'Ge°r.2

Further examples in Hawkins, Horae I39ff. For the tendency of the later synoptists and the MS tradition to make the same or similar corrections of Mk., see Turner,

Usage, passim.

6 8ε. Morgenthaler counts 160 instances of 8ε in Mk., 1078 of καί. In the quotations from the longer text there are 3 uses of 8ε and 18 of καί. Thus the proportions of δε to καί are 1 :6.j in canonical Mk. and 1 :6 in the longer text. This coincidence in proportion can scarcely be considered significant, since the quotations of the longer text are so short. However, its καί . . . καί . . . καί, with an occasional 8ε thrown in, is paralleled in many Markan stories : for example, that of Bartimaeus

(Mk. 10.46-52), which, according to Clement, immediately followed the second quotation. (See Hawkins, Horae i5off. on Mk.'s characteristic preference of καί to δε.) νεανίσκος. See above, on III. 3. Mark was fond of diminutives (Turner, Usage

29-349)· εμβλεφας αύτώ ηγάπησεν αυτόν, εμβλεφας/α at the beginning of a sentence, with a following finite verb: 3 in Mk., 3 in Mt. (1 Markan), and 1 Lk. In Mk. the sequence is twice εμβλεφας/α + verb. Two of the Markan usages (10.21,27) come close together, immediately before the place where the letter locates this usage

(following 10.34). This accords with Mark's habit of using the same construction several times in quick sequence. (See above, on κρατησας της χειρός, and, in Appendix

E, the distribution of καί ευθύς. [But other writers also have this habit, notably Paul and Luke. H.J.C.] αγαπάω is frequent in the Gospels, Mk. 5 (+ iL)), Mt. 8 (2

Markan), Lk. 13 (none Markan, 2 omitted by Dii.). Therefore, on stylistic grounds the fact that the clause εμβλεφας αύτώ ηγάπησεν αύτόν occurs in Mk. 10.21 is no reason to deny that Mark might have repeated it after 10.34. The question of the significance of the repetition must be postponed to the discussion of the content of the quotations from the longer text.

1 1 1

111.4-5

THE SECRET GOSPEL

111.5

/cat ηρζατο παρακαλεΐν αυτόν cva /xer' αυτου fj ηρξατοΙήρξαντο. With following infinitive: 26 in Mk. (D it. omit 3 but add 3 others),

9 in Mt. (6 Markan), 19 in Lk. (2 or 3 Markan) -fi 4 (2 Markan) in the D text of Lk. Turner ( Usage 28.352ff) and Hunkin (άρχομ at) remark that the use of ηρξατο/ντο with the present infinitive as a substitute for the imperfect is typically Markan.

/cat ήρξατο as the beginning of a sentence is particularly so: 10 in Mk., 3 in Lk., never in Mt. or Jn. (For comparisons with classical Greek and LXX see Doudna, Greek

5 iff and 1 1 iff.) παρακαλεΐν. Meaning "entreat," 9 in Mk., 6 in Mt. (4 Markan) + 1 D it., 5 in Lk. (3 Markan) of which 1 is omitted by XD al., but D it. also add an instance. παρακαλεΐν regularly governs the accusative. Followed by a ΐνα clause and the subjunctive: 5 or 6 in Mk., 1 in Mt. (Markan), 2 in Lk. (both Markan). LSJ cites the construction from Aristeas, Arrian, and others; in the Gospels it seems a Markan trait. Turner thought Mk. characterized by fondness for ΐνα and use of it with other than "its proper sense of purpose"; he included the uses after παρακαλεΐν in his long list of examples ( Usage 29.356). αυτόν . . . αυτόν. The repetition is particularly noticeable since it follows αυτώ . . . αυτόν in the preceding line. Doudna, Greek 36, remarks: "One of the outstanding features of the Greek of the gospels is the frequency of the oblique cases of the personal pronouns; Mark shares this to a slight degree" — and so, as he shows, do the papyri.

Kilpatrick, Atticism 136, takes the frequency of αυτός as one sign of a well-preserved, primitive text. ΐνα μετ' αυτόν f. Mk. regularly speaks of Jesus as acting μετά των μαθητών/τών δώδεκα (3·7 » 8. ι ο ; ι ι.ι ι ; 1 4* 1 4? 1 7) * But there seem traces of a tradition which spoke of the disciples as ol μετ' αύτοΰ. Thus the twelve are "made" ΐνα ώσιν μετ' αυτόν

(3.14)5 the Gerasene demoniac beseeches Jesus ΐνα μετ' αύτοΰ fj (5.18); when Jesus goes in to raise Jai'rus' daughter he takes with him only her parents and τούς μετ' αύτοΰ (5.40) ; when he leaves Capernaum he is pursued by Simon and ol μετ' αύτοΰ

(1.36, although, as H.J.C. notes, the reference here is not certain; Lohmeyer, ad loc., plausibly compares 16.7, τοΐς μαθηταΐς αύτοΰ καί Πετρω). Cf. also 2.19; 9*3^j

14.67. [Of the preceding passages, in 1.36; 3.14; 5.18; 5.40; 8.10, and 11.11, Mk.'s μετά is not paralleled in Mt. or Lk. R.S. is therefore justified in seeing here a Markan trait.] The phrase μετ' αύτοΰ with this sense appears also in Mk. 16.10 and Jn. 9.40

(cf. 6.66) and perhaps Lk. 22.59 (c^· Jn· 16*26). Further traces of the same usage may perhaps be seen in Mt. 12.30 and parallels; 26.51,69,71; 28.2o(?);

Lk. 22.21,28,33; 23.43; Jn. 13.8,33; 14.9; 15.27; 16.4; 17.12,24. Therefore, that ΐνα μετ' αύτοΰ fj should appear both in Mk. 5.18 and in the longer text is not surprising.

Moreover, we have here another near-coincidence with a false reading of D and its allies. In Mk. 5. 1 7f the great majority of texts read καί ή p ξαντό (sc. the Gerasenes)

1 12

THE SECRET GOSPEL

111.5-6 111.6 και βξζλθόντβς βκ τοΰ μνημείου ηλθον εις την οικίαν του νεανίσκου' παρακαλεΐν αυτόν απελθειν . . . καί . . . παρεκάλει αυτόν 6 δαιμονισθείς ίνα μ ετ' αύτοΰ η.

But D (with considerable support) reads παρεκάλουν αυτόν ίνα άπελθη . . . καί . . . ηρξατο παρακαλεΐν αυτόν 6 δα ιμονισθείς ίνα η μετ' αύτοΰ. This shift in the position of the ηρξα{ν)το construction could hardly have been motivated by considerations of style or meaning; it may reflect a memory of the longer text. Many D variants indicate that the D text of Mt. and Lk. was corrupted by the scribe's memories of canonical Mk. (e.g., those at Lk. 5.14 and 6.1; cf. below, on καί εξελθόντες, III. 5; μεθ * ημέρας, III. 6-7; etc.); similar corruption by memories of the longer text is therefore possible. [H.J.C., however, comments: "Memories of short Mk. affected scribes of Mt. and Lk., yes, but perhaps oftener memories of Mt. and Lk. affected scribes of short Mk."] καί εξελθόντες. Initial, with a following finite verb: 7 or 8 in Mk. (in 3 D it. have δε) ; 7 in Mt. + 1 in D, 1 in KBCL0, etc. Both these additional instances are Markan, but only one of the 7 secure readings is so; more contamination — see the preceding note. Only 2 in Lk., and of these 22.62 may be a gloss taken from Mt. Of the 8 in Mk., 3 are followed by εκ and 2 of these by έρχομαι. Of the certain instances in Mt.,

3 (and in B al. a fourth) are followed by εκ or its compounds, and one by εισέρχομαι.

Lk. has εξω after εξελθών in 22.62. καί εξελθόντες εκ τοΰ μνημείου ηλθον. Not found verbatim in the Gospels, but in Mt. 28.8 all uncials except KBCL(9, and most minuscules, have καί εξελθοΰσαι από τοΰ μνημείου', and Mk. 16.8 has καί εξελθοΰσαι εφυγον από τοΰ μνημείου (of the women after their Easter morning visit), while Mt. 27.53 has καί εξελθόντες εκ των μνημείων

. . . είσηλθον (of the dead raised along with Jesus). [P.B. compares also Mk. 1.29, where — after the cure of a demoniac — the text goes on, καί ευθύς εκ της συναγωγής εξελθόντες ηλθον εις την οικίαν Σίμωνοςό] In view of these multiple parallels it seems plausible to explain the text as free composition in conventional style rather than direct borrowing from any one passage, or word-by-word compilation from several.

[T.B. points out that the western text of Mt. 27.53, especially in the Syriac, is very close to the longer text of Mk. Sy.paL omits μετά, την εγερσιν αύτοΰ, DSy.palsLat.

( partim ) have ηλθον instead of είσηλθον, and the Syriac κ*ια=^3 canbe under stood as a singular. Perhaps the western variants may reflect some recollection of the longer text.] On εξελθόντες . . . ηλθον as a Markan usage see Turner, Commentary

155. On the subject of ηλθον, see above, on II.23, καί έρχονται. εις την οικίαν, οικία, Mk. 18, Mt. 25? Lk. 25* οίκος, Mk. 12, Mt. IO, Lk. 33· Again the longer text goes with Mt. and Mk. against Lk. References to particular houses which might be called "historical" (by contrast to those in parables or sayings) with οικία : 6 in Mk. (+ 2 in D), 9 or 10 in Mt. (only 3 Markan), 9 or 10 in Lk.

III. 6

THE SECRET GOSPEL

(+ i in D; only 2 Markan). Often these follow forms of (εις) έρχομαι with εις: 2 in Mk. + 2 in D, 5 in Mt. (1 Markan), 3 in Lk. (1 Markan.) This reference to going into a house often connects two scenes, as it does here: Mk. 1.29; 2.15; 9.33;

Mt. 9.23; 9.28; 13.36; 17.25. There is no reason to suppose this transition borrowed from any one of its many parallels. Kilpatrick, Notes 5ff, kindly put at my disposal an unpublished study in which he concludes that in Mk. οΐκος corresponds to "home," οικία to "house." He observes, however, that the two are sometimes interchangeable, as "home" and "house" in English. This would seem to be the case here. ην γάρ πλούσιος, ήν/ήσαν γάρ, as introduction of an appended explanation; 9 or 10 in Mk. (1 omitted by D it., etc., 1 by XB, etc.), 5 in Mt. (all Markan), 4 in Lk.

(+ i in D, only 1 Markan). It is a Markan trait (Turner, Usage 26.i45ff; Zerwick,

U nter sue hungen i3off; Bird, γάρ) and Mk. might have used it with πλούσιος as a predicate adjective. The canonical text of Mk. uses πλούσιος twice (a third use appears in a variant). However, πλούσιος is typical of Lk. (1 1 uses, against 2 in Mk. and 3 in Mt.), and rjv γάρ πλούσιος σφοδρά [Sy.0, omits σφοδρά — R.S.] appears in Lk. 18.23, where Mk. and Mt. have rjv γάρ εχων κτήματα πολλά. This is the young man in Mk. 10.22 of whom Mk. uses the phrase εμβλεφας αύτω ηγάπησεν αυτόν, also found in the longer text, above. Mk. 10.22 preceded by little the present passage of the longer text, which stood after 10.34. Given Mark's habit of repeating himself after short intervals, he probably would have repeated εχων κτήματα πολλά, rather than have summarized it with πλούσιος. The longer text here may have been cor rupted by a copyist's memory of Lk. or by a gloss. Perhaps the latter is more likely, since the rest of the text shows no knowledge of Lk. (except in the concluding clause of the second quotation, which is probably spurious; see below, on III. 16). Yet another possibility is that Clement misquoted; for the contamination of his quotations of Mk. by his memories of Mt. and Lk., see Appendix L. On the other hand, the dangling phrase is in Mark's manner and makes sense: They went to the young man's house, for he was wealthy (and therefore able to receive the company which traveled with Jesus). Cf. Zacchaeus, Lk. 19. iff. [W.M.C. remarks that the subjects of εξελθόντες . . . ηλθον εις την οικίαν are, strictly, only Jesus and the youth.] However, as Turner noted ( Usage 26.231), Markan reports of the movements of Jesus, even when the verb is in the singular, are normally to be understood as implying that his followers went with him. See the note on και έρχονται, in II. 23, above. The notion that πλούσιος is a copyist's corruption can be supported by the fact that a similar corruption has introduced πλούσιος into Mk. 10.17 in AKMW0/7, many minuscules, and some MSS of the Old Latin, Syriac, Coptic, Armenian, and Georgian translations. [My attention was called to the variant by R.S.] Turner, Readings 6, thought the κτήματα in Mk. 10.22 a corruption introduced from Mt. 19.22 and maintained that the correct reading was χρήματα, preserved by the western text and Clement and repeated — as usual — by Mk. (in the following verse, 10.23).

111.6-7

THE SECRET GOSPEL

III.7 καί μεθ' ημέρας εξ επεταξεν αντω ο ' Ιησούς * καί όφίας γενομενης ερχεται καί μεθ' ημέρας έξ. Mk. is characterized by its separation of events from eath other by precise numbers of days, intervals it indicates by μετά with the accusative: 5 instances. Mt. has only 3 (2 Markan and 27.63 — probably an echo of Mk.); Lk. has only 2 (and 9.28 seems a deliberate correction of Mk.). DzY. have 2 more instances in Mt. and one more in Lk. (another example of the corruption of their archetype by the influence of Mk. ; see above, on μετ αύτοΰ). Per contra, the Matthaean usage

(with the dative) has corrupted the later uncials of Mk. If Lk. 9.28 is a deliberate correction of Mk. 9.2 (so Rengstorf on Lk. ad loc.), there was some tradition about these intervals and some importance attached to them. Therefore the specification of the interval may be a datum of tradition, not an echo of Mk. 9.2, where the same words occur. On Mark's fondness for giving specific numbers, Turner, Usage 26.337flf. έττέταξεν. The verb: 4 in Mk., never in Mt., 4 in Lk. (1 Markan) + 1 in D. The person commanded is always in the dative. The form έπέταξεν occurs twice in Mk. and in the D variant to Lk. (8.55). έπέταξεν 6 ' Ιησούς αύτοΐς is found in D it. to Mk.

6.39 (where other witnesses lack ό Ίησοΰς). These parallels demonstrate merely that the word was used normally by Mk. and Lk. The peculiarity here is the failure to specify the content of Jesus' command; that is understood from the context, as in Mk. 1.27; Lk. 4.36; 8 25. [C.F.D.M., however, remarks that έπέταξεν αντω without direct object is odd, and the parallels adduced here are not quite similar for in all of them the content of the verb is perfectly clear. Moreover, why did the young man have to come to Jesus and stay with him, if Jesus was at his house?] The direct object may have been part of the secret oral teaching. It will be argued later that the young man came to Jesus to receive baptism, conceived as a magically efficacious rite. If so, he had to come to Jesus becauses Jesus had to prepare (purify? exorcise?) the area and the materials for the rite. The story suggests a large house, perhaps a villa. The young man was rich. Jesus and his followers may have been given awing for themselves. καί όφίας γενομενης. όφίας γενομενης as a genitive absolute at the beginning of a sentence: 4 in Mk. (3 introduced by /cat), 6 or 7 in Mt. (6 with 8έ, never with καί), never in Lk. or Jn. It is followed by έρχεται in Mk. 14.17 (the introduction to the last supper — a secret ceremony, like the one here, but this may be chance), by ήλθεν in Mt. 27.57 (Joseph of Arimathea), έλθών in Mk. i5-42f (also Joseph of Arimathea), and προσηλθον in Mt. 14.15 (the introduction to the feeding of the five thousand).

By contrast with these stylistic affiliations, the content of this story resembles Jn. 3:

Nicodemus ηλθεν ττρός αυτόν νυκτός and received instruction on baptism as necessary for those who would enter the Kingdom of God. έρχεται. As a historical present: 10 in Mk. (+ 1 in X*A, etc., and 2 in D it., etc.),

4 in Mt. (2 Markan), 1 in Lk. || έρχονται, historical present, in Mk.). Recognized as a distinctively Markan trait by Hawkins (

Horae 143). It is so frequent that its con junction with the also frequent καί όφίας γενομενης is no evidence for dependence on Mk. 14.17.

THE SECRET GOSPEL

III. 8 συν αυτω την νύκτα 6 νεανίσκος. See above, on III. 3. This is the fourth occurrence of the word in 6 lines. More Markan repetition. προς αυτόν, πρός after έρχομαι : 12 in Mk. ( + i in D &it.) ; 12 in Mt. (+ 1 dubious);

9 in Lk. Only 2 of the Matthaean and 2 of the Lucan constructions are Markan.

Therefore it would at first sight seem pure chance that πρός αυτόν after έρχομαι occurs 7 or 8 times in Mk., once in Lk., and never in Mt. But προσέρχομαι was used

52 times by Mt., 5 by Mk., and 10 by Lk. περιβεβλη μένος σινδόνα επί γυμνού. Verbatim in Mk. 1 4.51 (except that W fam. i.r.LSy.sCop.sa· have discreetly omitted επί γυμνού , while Θ fam.i^[exc. 124]. 543,

565.Sy.p Eth. have accidentally replaced it by the γυμνός of the following verse). In 14.51, too, the subject is νεανίσκος τις — the young man in a sheet who was with Jesus at the time of his arrest and who, on being seized, fled naked (an episode both Mt. and Lk. chose to omit). All the words in the phrase (except επί) are comparatively rare in the synoptics: περιβάλλειν , Mk. 2, Mt. 5, Lk. 2; σινδών, Mk. 4, Mt. 1, Lk. 1 ; γυμνός, Mk. 2, Mt. 4, Lk. o. Consequently, the occurrence of the phrase both in the longer and in the canonical texts of Mk. can hardly be explained as an accident of free composition. Either the phrase was a fixed formula in the life of some early church (a baptismal rubric?) or its presence in both texts is evidence of some historical connection. εμεινε. 2 in Mk., 3 in Mt., 7 in Lk., 40 in Jn. Of the 3 in Mt., 2 are Markan; of the 7 in Lk., 1. εμεινεν δε . . συν αύτη ώς μήνας τρεις, in Lk. 1.5b (the visitation), καί παρ' αύτω εμειναν την ημέραν εκείνην, in Jn. I ·39 (the first disciples), καί εμεινεν εκεί δυο ημέρας, in Jn. 4·4° (Jesus in Samaria; cf. 10. 40; 1 1.6), and κακει εμεινεν μετά των μαθητών in Jn. 1 1.54 (after the raising of Lazarus) testify to common usage rather than literary dependence. (Some historical fact of Jesus' practice may underlie the similarity with the story of the first disciples; or the basis may be some early church usage, or mere chance.) μένω was affected byjn. because of its theological connotations (e.g., 1 5.4fT) , which probably had some connection with baptism and might be relevant here; but the word is too common to justify speculation. συν. To indicate the person accompanying or accompanied: 5 in Mk., 2 in Mt.,

23 in Lk. Cf. πρός αυτόν in the preceding line. Another instance of Mt.'s preference for compound verbs. νύκτα. 4 in Mk., 9 in Mt., 7 in Lk. + 1 D it. (11.30). In the accusative of extent of time: 3 in Mt. and Lk. 11.30. Cf. also the adverbial accusatives νύκτα καί ημέραν

1 16

THE SECRET GOSPEL

III. 9-10

6eov. in Mk. 4-27 and Lk. 2.37? and τάς 8ε νύκτας in Lk. 21.37. These parallels are insig nificant; the longer text probably derived its phrase from common usage, not from a literary source. e κείνην. The position after the noun is normal in Mk., as shown by Kilpatrick, εκείνος. ε8ί8ασκε γαρ αυτόν. 8ι8άσκειν : Mk. 1 7, Mt. 1 4, Lk. 1 7. The form ε8ί8ασκεν used of Jesus: 6 in Mk., 2 in Mt. (neither Markan), 2 (+ 1 D) in Lk. (none Markan). In Mk. 9.31 it is followed by γάρ and refers to the secret teaching of the passion. This follows the stories of the transfiguration and the raising of the demoniac boy, both of which have similarities to the story in the letter's Gospel. Note especially the se quence 9.27 ό 8ε Ίησοΰς κρατησας της χειρος αύτοΰ ηγειρεν αυτόν . . . (.

3 °). κάκεΐθεν εξελθόντες . . . ούκ ηθελεν ΐνα τις γνοΐ (31), ε8ί8ασκεν γάρ τούς μα θήτας αύτοΰ , etc. Mk.

9.27-3 1 precedes by little the place (10.34) where the letter locates its Gospel frag ment, and the recurrence of similar constructions after short intervals is typical of Markan style. As to content, on the other hand, the synoptics never represent Jesus as teaching a single person; but accounts of his having done so are prominent in Jn.

(3, Nicodemus; 4, the woman of Samaria; etc.). Kilpatrick, Mission i49ff, points out that Mk. frequently uses verbs to begin sentences. [W.M.G. here finds "the interlaced word order (verb — outer object — subject— inner object) . . . contrived."] It is not precisely paralleled in Mk. in sentences of which Jesus is the subject, but Mk. often has the similar order: verb — indirect object — subject — direct object — 1.25; 2.19;

6.4; 12.24; 14.27,30. In all these, as in the longer text, the verbs imply speaking, the first object is the person spoken to, Jesus is the speaker, and the second object is the thing said. το μυστήριον της βασιλείας του θεοΰ. μυστήριον is used only once in each of the synoptics, and in both Mt. and Lk. its use is a parallel to Mk. 4.1 1, ύμΐν το μυστήριον

8ε8οται της βασιλείας του θεοΰ (+ γνώναι , C2Dzh9T0b[ exc. Κ] fam.l.22.fam.l^.^^.

28*33· 1 57·5^5·597·700· 1 °7 1 ml. pier. it.vg.SY.peshMCop.h0-aliq-Geo.Aeth. Arm. ; τά μυσ τήρια , ΟΣΦ fam.i. 67.106.1 15.201. 235*258.5I7*5®9*Sy-hl'Arm.). Cf. Mt. 13.11, ύμΐν

8ε8οται γνώναι τά μυστήρια της βασιλείας των ουρανών ( mysterium a.C.d.f. ff2.g1.l.q.aur.Vg.

[3 MSS.]Sy.c,t,pesh Geo.Aug.serm·165 ; sacramentum k); Lk. 8.10, ύμΐν 8ε8οται γνώναι τά μυστήρια της βασιλείας τοΰ θεοΰ ( γνώναι transp. post θεοΰ D, om. <2; το μυστήριον Cit.vg.

Sy.Clem.Iren.). The uses of το μυστήριον in MSS of Mt. are further examples of the early corruption of the western text by reminiscences of Mk. Matthew's interpretation of the Markan phrase, by changing το μυστήριον to τά μυστήρια and adding γνώναι,

III. io— 1 1

THE SECRET GOSPEL

III. 1 1 €K€ Wev 8e άναστας επέστρεψβν els το πέραν του * Ιορδάνου . got into Lk. (where the secondary character of γνώναι is indicated by the difference as to its position) and almost got into Mk. The use of το μυστήριον in the longer text of Mk. is either a Markan trait or evidence that το μυστήριον της βασιλείας του θεού was a fixed phrase in some circles of early Christianity. Its significance will be dis cussed below, in the section on content. [W.M.C. remarks that we never hear what happened to the youth's sister.] Such disappearance of minor characters is typical of Markan narratives: so Simon, Andrew, James, and John in i.4off; the men who brought the paralytic in 2.3ff; Levi in 2.i5ff; etc. εκεΐθεν. 5 in Mk. (and κάκείθεν , in 9. 30), 12 in Mt., 3 in Lk. (and κάκεΐθεν, in 1 1.53).

With a participle, as the beginning of a sentence, 3 in Mk. (+ 2 more in D), 5 in Mt. (only 1 Markan), 1 in Lk. In Mt. the participle always precedes and is never άναστάς. In two of the well attested uses in Mk. (7.24 and 10.1) the participle is άναστάς and follows εκεΐθεν, as διαπεράσαντες in the western text (Dz7.) of Mk. 6.53, and εξελθόντες after κάκεΐθεν in 9. 30. The usage here is characteristically Markan.

The exact phrase εκεΐθεν δε άναστάς appears in Mk. 7.24; IO.I has και εκεΐθεν άναστάς

[and R.S. remarks that it is followed by ερχεται εις . . . πέραν του 'Ιορδάνου]. άναστάς. The participle used pleonastically, imitating the LXX translation of Dp, with a following finite verb (Blass-Debrunner-Funk, no. 419.2); 4 or 6 in Mk., 2 in Mt. (both Markan), 11 or 12 in Lk. (only 1 Markan). Certainly not a Matthaean usage. See above, on II. 24, καί + participle. επεστρεφεν. The verb: 4 in Mk. (1 with εις, 13.16), 4 in Mt. + 2 D, 7 in Lk. + 1 D

(with είς, 2.39; 17.31 || Mk. 13.16). The usage is normal. εις τδ πέραν του 'Ιορδάνου. Not found in the synoptics. Mk. 5.1 has εις τδ πέραν της θαλάσσης; Lk. 8.22, εις το πέραν της λίμνης; the Alexandrian text of Mk. IO.I, διά τοΰ πέραν του ' Ιορδάνου. Otherwise the usage is either το πέραν with no following genitive

(Mk. 4.35; 5.21 ; 6.45; 8.13 and parallels), or πέραν τοΰ Ιορδάνου without the article

(Mk. 3.8; 10.1 — see variants in Appendix E). These seem to represent the normal

Markan usage, and the expression is predominantly Markan: Lk. has it only once, where he takes it from Mk. and modifies it; Mt. has 7 uses, all Markan except 4.15, a quotation from LXX Is. 8.23. Accordingly it might be argued that the words τοΰ

' Ιορδάνου in the letter's Gospel are an epexegetic gloss. But all the Markan uses of το πέραν alone refer to the opposite side of the Sea of Galilee and are explained by the context, while in the letter's Gospel the determinant τοΰ Ιορδάνου is required by the context. Mk. 5.1 is a parallel case.

1 18

III. 14-15

THE SECRET GOSPEL

With the above words the first of the letter's quotations from the secret Gospel concludes. There follow the comments : επί μεν τουτοις επεταυ τό, " καί προσπορευονται αντίο ' Ιάκωβος καί 'Ιωάννης ," <(Mk. 10.35)* Ka'L πάσα η περικοπή, τό Se " γυμνός γνμνω " καί τάλλα περί ών εγραφας ούχ εΰρίσκεταν μετά Se τό, (i καί ερχεται είς ' Ιεριχώ " (Mk.

10.46a) επάγει μόνον (III. 1 1 — 1 4) * These have been discussed in the commentary on the letter in Chapter Two, above. The letter's second quotation from the secret Gospel reads as follows :

ΠΙ· 15 καί ήσαν εκεί η αδελφή του νεανίσκου ον ηγάπα αυτόν 6 Ίησοΰς καί ησαν εκεί. See above, on καί rjv εκεί, in II. 23. For the plural the closest parallels are Mk. 2.6, ησαν Se τινες των γραμματέων εκεί, where Sy.pesh,hL perhaps read εκεί before τινες, and Mt. 27.55 Vaav &ε εκεί γυναίκες where SyAAeth.Geor.1 put the conjunction at the beginning. Such phrases are narrative cliches and can hardly serve as evidence of dependence. [T.B. queries: Is it coincidence that in Mt. 27.55 the phrase also introduces a group of women, and that this group also — to judge from the parallel in Mk. 15.40 — once contained Salome?] It is not impossible that Matthew may have used the longer text of Mk. Cf. the remarks on προσεκυνησε, above, II. 24. ή aSeX φη του νεανίσκου. Unparalleled. Sisters of a specific brother (other than Jesus) appear only in Jn. 11 — the Lazarus story — and νεανίσκος only in the synoptics; see above. ον "ηγάπα αυτόν 6 Ίησοΰς. See above, on εμβλεφας, in III. 4· Although all the Gospels use αγαπάω frequently, the synoptics only once speak of Jesus' loving anybody — the man who questioned him in Mk. IO.21 : o Se Ίησοΰς εμβλεφας αύτω ηγάπησεν αυτόν.

Jn· says Jesus loved (ηγάπα) Martha and her sister and Lazarus (11.5) and his dis ciples (13.1,34; 15.9,12) and the Father (14.31) and an unnamed disciple ον ηγάπα ό Ίησοΰς, thus referred to 4 times: 13.23 (in the last supper); 19.26 (at the cross);

21.7,20 (the resurrection appearances at the Sea of Galilee). Cf. the άλλος μαθητης of 1 8. 1 5f and 20.2ff. It has often been argued from Johannine evidence that the un named "beloved disciple" was Lazarus (e.g., recently, Eckhardt, Tod 11—20). The longer text strengthens the argument by first telling a version of the Lazarus story in which the dead youth's sister plays an important role and the youth is said to have loved Jesus, and then locating shortly after this a reference to the disciple whom

Jesus loved and his sister. So the Markan and the Johannine traditions here, as often, are remotely similar. Therefore it is not surprising that the same traditional formulas should occur in both, ον ηγάπα ό Ίησοΰς is a fixed periphrasis in Jn. and the reader will immediately suppose that the longer text got it from Jn. But then the reader will have to explain why the longer text shows no other trace of John's peculiar phraseology. Matters of plot are not in question here ; they will be dealt with in the

III. 1 5—1 6

THE SECRET GOSPEL

III.16

/cat yj μητηρ αύτοΰ καί Σαλώμη, καί ουκ άπβδβξα το αυτά? 6 Ίησοΰς. next chapter. The present question is: When an isolated Johannine phrase occurs in a text which otherwise shows no important traces of Johannine phraseology, is the isolated phrase to be taken as proof of dependence, or is the absence of other traits to be taken as proof of independence? Are we to prove the dependence of ζ) on Jn. by Mt. n.25f II Lk. io.2if? [H.J.C. compares also the Johannine ούδε 6 υιός, cl μή

6 πατήρ , in Mk. 13.32 and the parallel between Jn. 3.3-5 and Justin, First Apology

61.4-5.] Notice also the similarities to John in the phraseology of the Dead Sea documents (Brown, Schrolls), and the phrases of canonical Mk. which appear in section IV of Appendix G with parallels only from Jn. Evidently, a number of phrases best known to us as "Johannine" were taken by John from his Christian and Jewish environment. One of these phrases seems to have been the periphrasis ον ήγάπα 6

* Ιησούς , which apparently was a fixed formula in at least two strains of early Christian tradition, like (o) cl ς των δώδβκα as a designation of Judas: Mk. 14.10,20,43; Mt.

26.14,47; Lk. 22.47; Jn· 6-7 1 (+ eV; cf., however, 20.24, where the same formula refers to Thomas). An early date for the periphrasis is suggested by its anonymity.

(Cf. Bultmann, Geschichte 72, 256^ etc. Turner's argument to the contrary — Usage

26.338 — from the practice of rhetoricians is irrelevant. The Gospels were not written by rhetoricians and their content shows the authors did multiply names.) Further evidence that the longer text did not get its formula from Jn. appears in the pleonastic αυτόν , to which the uses in Jn. afford no parallel, and which a writer familiar with Greek would hardly have added; it is probably a Semitism — cf. ής . . . αυτής in II. 23, above, and the note there. [P.B. would distinguish the examples of this construction in the longer text and in Mt., where he thinks them Semitisms, from those in canonical

Mk., where he thinks them emphatic, and would find in this distinction evidence that the letter's Gospel is not by Mark.] The distinction seems to me so fine as to be subjective; it escaped Moule, Idiom-Book 176, and Blass-Debrunner-Funk no. 297. ή μήτηρ αύτοΰ. Verbatim in Mk. 3.31 ; Mt. 13.35; Lk. 1,60; 2.48,51 ; 8.19 (XDzfiSy.) ;

Jn. 2-5,· 2; 19.25. The phrase is standard and cannot be referred with confidence to any single source.

Σαλώμη. Salome appears in the NT only in Mk. 15.40 and 16.1, in both as the final figure in a list of female witnesses of an important occasion (15.40, the cruci fixion; 1 6. i, the discovery of the empty tomb). In 15.40: Μαρία ή Μαγδαληνή καί

Μαρία ή ' Ιακώβου του μικρού καί Ίωσήτος μήτηρ καί Σαλώμη. In 16.Ι : Μαρία ή Μαγδαληνή καί Μαρία ή

' Ιακώβου καί Σαλώμη. (There is considerable difference between MSS as to the spelling of the second Mary's name and her connection with James and Joses.) The list in the longer text of Mk. is of the same type as the other two, and may be of the same women. Luke (23.55; 24.1) has omitted both the lists pre served in canonical Mk. ; Matthew has deleted the name of Salome from the first list

THE SECRET GOSPEL

III.16

(27.56) and deleted her altogether from the second (27.61; 28.1); John (19.25) has replaced her by "the sister of his (Jesus') mother" (or?) "Mary of Klopas." [R.S. remarks that if we suppose the lists to be of the same women, and if we suppose the beloved νεανίσκος here to be the same as the loving νεανίσκος in the former fragment, and if we therefore suppose him to be Lazarus, then his sister here would be the Mary

Magdalene of the other Markan lists. And since Mary, Lazarus' sister, anointed Jesus in Bethany in Jn. 12. iff (cf. Mk. 14.3; Mt. 26.27), and a woman who was a sinner anointed Jesus in Lk. 7.36, it would follow that Mary Magdalene, Lazarus' sister, was the sinner <(from whom also seven devils were driven out, Lk. 8.2 — she had an eventful life.) Again, Mary, the mother of James and Joses, of the other Markan lists, would be "his mother" in the list in the longer text, and "his" would be Jesus', since James and Joses were his brothers: Mk. 6.3, ονχ οΰτός εστιν ό τεκτων 6 νιος της

Μαρίας καί αδελφός ' Ιακώβου καί Ίωσητος καί ^ Ιούδα καί Σίμωνος.\ If this latter iden tification be not accepted, Mk. did not mention the mother of Jesus among those· who witnessed his passion and burial and discovered his resurrection. On the other hand, by similar reasoning Salome would be both "the mother of the sons of Zebedee," who replaces her in Mt. 27.56, and "the sister of (Jesus') mother" (or?) "Mary of Klopas," who replaces her in Jn. 19.25. [In favor of at least the former of these identifications, R.S. points out that "the mother of the sons of Zebedee" appears in Mt. 20.20, where she may again be a Matthaean substitute for Salome, who played a similar role in the following story of the longer Markan text. See above, on II.24.]

The role of Salome in early Christian polemics will be discussed later in this chapter

(section III. D.4, "evidence for abbreviation at mk. 10.46"). άπεδεξατο. In the NT, αποδέχομαι is found only in Lk.-Acts. Lk. 8.40 (άπεδεξατο), of the crowd's giving a good reception to Jesus; 9.1 1, of Jesus' receiving the crowd kindly. 5 uses in Acts, of which 3 (18.27; 2 ί . ϊ 7 ; 28.30) have the same sense (receive a person kindly) and the same construction (an immediately following accusative of the person received). [With the whole phrase here, C.F.D.M. compares Lk. 9-53? καί ούκ εδε ξαντό αυτόν, of the Samaritans' refusal to receive Jesus.] This clearly Lucan trait contrasts as such with the preceding text, which is almost entirely free of Lucan traits (the only very probable one being πλούσιος, above, III. 6). Clement uses the verb often in the cognate sense, "approve of" a person — that is, of what he says or does — L178.27; 186.8; 223.10; 265.3; II.3.8; 4.13; 235.25; 237.21. A number of these citations, particularly the last two, show the disciplinary connotation — almost

"accept as communicants" — which the word has in Clement's usage. It is this con notation which the word is meant to carry in the above text of the letter's quotation, and which is necessary to give force to the otherwise trivial ending. The story, as Clement quotes it, is quite unlike any other NT story because it has no apparently significant content. There is no miracle, no saying, nothing but Jesus refusal to receive, on one occasion, three women. Therefore the story, as it stands, can have been invented and preserved only as polemic against these women or their followers or persons who appealed to their authority (as the Carpocratians did to that of Salome: Origen, Contra Celsum V.62). But that such a bare polemic notice was what

III. 1 6

THE SECRET GOSPEL

originally stood in the longer text is almost incredible : it is too little like the patterns of Gospel stories. The original text must have gone on to report some action or saying of Jesus. Accordingly, this final phrase with its sudden change of vocabulary and its anachronistic church discipline is to be attributed to Clement or some editor of almost Clement's time and vocabulary who deleted the original ending of the story as it stood in the longer text of Mk. and substituted this Lucan phrase (with its second-century meaning) for what he had deleted. What he had deleted we can only guess from the preceding context and from the following — and last preserved — sentence of the letter: τα δε άλλα τα πολλά ά εγραφας φευσματα καί φαίνεται καί εστιν.

Most likely it was a conversation with Salome (again see below, in section III).

B. Synthesis of findings

I. INFLUENCE ON THE WESTERN TEXT

Perhaps the most surprising of the facts revealed by the above commentary is the evidence for the influence of the longer text of Mk. on the western text of the canonical

Gospels (or, in Klijn's terminology— Survey, 4 — the "western readings" of the fathers and the Gospel MSS). The relationship of the two texts was independently discovered by Prof. R. Schippers and myself. There are three cases in which an apparently wrong reading in the western text affords the closest parallel to an apparently correct reading in the longer text of Mk: καί έρχονται εις Βηθ ανίαν II. 235 καί οργισθείς II. 25 5 and καί ηρζατο παρακαλεΐν αυτόν 'ίνα μετ' αύτοΰ η III .4 — 5 * ^et °ther cases appear in III. 4 εζετεινεν την χεΐρα καί ηγειρεν αυτόν and III. 7 επεταξεν αύτω 6 Ίησοϋς, but in III. 4 the parallels are so loose and in III. 7 the wording is so banal that neither constitutes evidence. By contrast, the error of Βηθ ανίαν in Mk. 8.22 and the difficulty of οργισθείς in i. 4 1 demand explanation, while the extent of the paralleled phrase in III. 4-5 makes the supposition of influence notunlikely (but does not prove it; see above on III. 1-2). The conjunction of the three — not to say five — cases is impressive. As to the question, Which influenced which ? — it is surely more probable that the bad readings of the western text were produced by contamination from the longer text, than that the good readings of the longer text were produced by a selec tion of the errors peculiar to the western. As already remarked, contamination is characteristic of the western text (Williams, Alterations if; Glasson, Mt. i8of), and contamination of MSS of the canonical Gospels by uncanonical material has often been demonstrated (Black, Aramaic 204 and 214; Williams, Alterations 7, i2ff, 23).

Qpispel ( Thomas 198; Hebraer euangelium 142), has argued that the western text was influenced by the Gospel according to the Hebrews ; a fortiori, it might have been in fluenced by the longer text of Mk. Further, the longer text shows a number of readings which are nonwestern: γυνή ης . . . αυτής . . . ελθοΰσα (II. 23), ηγειρεν αυτόν κρατησας της χειρός (III. 3-4); καί ευθυς (II. 26-in. ι ; III. 2). If the longer text be supposed to derive from the western, these nonwestern readings can hardly be accounted for.

Therefore the more probable explanation of the facts would seem to be that the

THE SECRET GOSPEL

archetype of the western text was contaminated by the scribe's recollections of the longer one. [But the influence may have been more contemporary than "recollections " implies. H.J.G.] This explanation accords with the fact that Clement both quoted the longer text and used the western (Barnard; Klijn, P. Bodmer //, 327, 332, 334).

[Moreover, as Pierson Parker pointed out to me, if Clement did not use the western text of Mk— a possibility which Swanson, Text 102, leaves open — then the western readings in the longer text cannot be corruptions due to Clement, but must come from an earlier tradition.] The theory that the western text was shaped by the longer text of Mk. also accords with the date reached above for the longer text (before 125) and with the current dating of the western text, which is now thought to have originated in Egypt about a.d. 150 (Duplacy, Ou en est I.43of). It also suggests both that the scribe who produced the archetype of the western text regarded the longer text of Mk. much as he did the canonical Gospels, of which he also conflated the readings, and that the sections of the longer text quoted by Clement had some im portant role in the life of the scribe's church, since so many recollections of them turn up in his text. The importance of the sections quoted by Clement might have been inferred also from the facts that they played a leading part in the argument between Theodore and the Carpocratians and that Clement chose to reassure Theo dore particularly about them. To the question of what their importance was we shall return later.

Postscript: Further evidence for these conclusions is now afforded by the addition to the Latin translation of Origen's commentary on Mt., to which Klijn, Question , has called attention. It quotes from a Gospel according to the Hebrews a text closely related to Mt. i g. 1 6ff but deviating from it in details which agree with or approximate the western text (Sy. and Diatessaron ). Klijn (154O thinks these details would indicate its priority to the Diatessaron , but he then rejects this conclusion as "inconceivable."

Now Mt. i g. 1 6ff is the synoptic parallel to Mk. io.iyff— the story of the "rich young ruler" which stood shortly before Clement's quotation from the longer text and was closely connected with it, as will be shown below. It is just possible, therefore, that this fragment may be another scrap (more or less rewritten) of the longer text. The close relation to Mt. and the attribution to the Gospel according to the Hebrews both fit.

2. VOCABULARY, PHRASEOLOGY, AND GRAMMAR

This establishment of the dependence of the western text on the longer text of Mk. enables us to rule out further consideration of the parallels peculiar to the western text. Turning now to the relation between the longer text of Mk. and the Nestle-Kilpatrick text of the canonical Gospels, we observe that the parallels pointed out in the above commentary are of two sorts:

On the one hand there are many brief parallels of words or phrases, like /cat

έρχονται et<r, καί e/cet, and so on, which are most frequent in one or anothei of the synoptics — usually Mk., occasionally Mt., very rarely Lk. or Jn. — but which are so commonplace that they cannot be used as evidence of dependence on any particular

THE SECRET GOSPEL

passage. Perhaps it is not strictly impossible that a compiler should have picked them out from remote parts of different Gospels and pieced them together in a mosaic

(where, nevertheless, there are no abnormal joints, no irrelevant words, no clear signs of mosaic composition!). But it is more plausible to suppose them the result of free composition by an author to whom the formulas of Markan style came as easily as they did to "Mark" himself. (Gf. the similar judgments of Dodd, Historical Tradi tion , passim.) On the other hand there are several exact parallels of considerable phrases. Con spicuous examples are vie Δαβίδ βλβησόν μβ, πβριβββλη μόνος σινδόνα βπί γυμνόν. These cannot be explained as accidental results of free composition in a vocabulary full of fixed formulas. Either they must be evidence of literary relationship to the passages where they stand in the canonical Gospels, or they must come from the technical terminology and fixed tradition of early Christianity. In either case they have some special theological significance. Consequently, their relation to their parallels in the canonical Gospels must be thought meaningful. (Contrast the cliches like καί

'έρχονται βις, of which the use in one or another story is a mere matter of chance, signifying nothing.) Unfortunately, the dividing line between these two groups is not clear, όμβλόψας αύτω ηγάπησβν αυτόν almost certainly belongs to the latter, άπβκυλισβν τον λίθον and από της θυρας του μνημβίου doubtless result from assimilation of the Lazarus story to the Easter story, or vice versa, but just when the assimilation took place is hard to say. rjv γαρ πλούσιος may be anything from ancient accident to medieval corruption.

The γυνή ης . . . αυτής . . . βλθοΰσα προσβκυνησβ sequence is probably a mere collection of cliches of which the arrangement was determined by content. And other cases suggest yet other relationships.

Consequently, we shall first collect the evidence afforded by the vocabulary, the phraseology, and those grammatical traits which have been picked out as charac teristic of one or another of the evangelists. This sort of evidence is the most objective, it will include the shorter parallels, and it may yield some results which will be help ful in the more difficult questions raised by the longer parallels.

The bulk of the vocabulary is made up of common words, and the different num bers of their occurrences in the several Gospels usually reflect the different sizes of the Gospels. The fragments of the longer text are so short that arguments from the numbers of times it uses words are mostly worthless ( βυθός and perhaps καί and δβ may be exceptions: see above, on III. i and 4). The distribution of νβανίσκος is remarkable but accords with Mark's habit of rapid repetition; cf. the distribution of διαστβλλομαι (4 from Mk. 7.36-9.1, i in Mk. 5.43, i in Mt., none in Lk. or Jn.), κβντνρίων (3 from Mk. 15.39-45» none elsewhere), κράββατος (4 in Mk. 2.4-12, 1 in Mk. 6.55, none in Mt. or Lk., 4 in Jn.), and so on.

That 12 words from the longer text are not in Jn., as against 3 not in Mk., 4 not in

Mt. and 3 not in Lk., is further evidence of the remoteness of the longer text from Jn. Mk. io.26b-34 (which, henceforth, we shall call CS — canonical sample ) has 4 not found elsewhere in Mk. ( αδύνατος , βκατονταπλασίων, μαστιγόω , and συμβαίνω) , 3 not in Mt., 3 not in Lk., and 14 not in Jn.; the similarity of these two sets of figures

Mt.

Lk. a. Vocabulary*

SG Mk.

Mt.

Lk.

SG Mk.

Jn. από I

M55

I — — I

I I

it I

I — δ e

r ( /

1 52 I —

ου I 36 1

I I

I I

I Η I I I

I I I

— — — > / 67 TL I

I I I

* Figures from canonical Gospels from Morgen thaler; for longer text of Mk. see Index I, s.v. Secret Gospel ( SG ). f Two more uses of γυμνός appeared in the Carpocratian version ( γυμνός γυμνω).

Vocabulary (figures from Morgenthaler) of the 1 75 words of canonical Mk. preced

Lk. ing the first quotation from the longer text (i.e., Mk. io.26b-34).

Mk.

Mt.

Lk. ιΊ Mk.

Mt.

— I I

1 17

— Jn·

I I

i74

I I

l6 ό, ή, τό 19 :5°4 2777 2629

— tf e/ f/

— —

"> / *5 εαν I

36 1 442 *37 προάγω I

— — ΤΙ? I

f / ζωτ} I

uio? I

ή' 6

t The numerals following each word indicate the number of its occurrences in Mk.io.26b~34.

THE SECRET GOSPEL

is striking. The occurrence in the longer text of a good number of John's favorite words should not be made an argument for relationship. John is notorious for the frequency with which he uses his favorite words, of which the longer text of Mk. contains, αγαπάω , αποθνήσκω, εγώ, ειμί, εκ, εκείνος, έρχομαι, 'Ιησούς, ΐνα, μαθητής, μένω, μνημεΐον, οπού, ου, and προσκυνάω. It would be absurd to suppose that the uses of these common words in the other Gospels were signs of Johannine influence. GS has the 5 underlined above, plus αιώνιος, αλλά, άμήν, ζωή, λαμβάνω, νυν, on, ούδείς, πάλιν, πατήρ, and Πάτρος. Similar collections can be made for Lk. (Longer text, γίγνομαι, ήμερα, θεός, ος, πλούσιος, πρός, συν, φωνή', GS, ημείς, ήμερα, θεός, νυν, ός, ούτος ) and Mt. (Longer text, γάρ, εγείρω, εκεΐθεν, εκείνος, ' Ιησούς, προσέρχομαι, προσκυνεΐν ', GS, άμήν, γά ρ, ή,

Of Mk.'s favorite words (as listed by Morgenthaler, 181, on grounds of frequency only) the longer text contains ευθύς, ΐνα, κρατεω, and όπου', GS, άλλά, δύναμαι, δώδεκα, πάλιν, and πολύς. The brevity of these lists, by comparison with those from the other

Gospels, is explained by the small number of "favorite words" to be found in Mk.

Mk. has the largest vocabulary in proportion to its size of any of the Gospels: 1,345 words in 11,200. Mt. has 1,691 in 18,300, Lk. has 2,055 in 19,400, and Jn. has only

1,011 in 15,400: (Morgenthaler, 164). Therefore Mk. uses most words least often,

Jn. least words most often. Accordingly, the list of Jn.'s favorite words contains 75 items, and any NT text is sure to show a substantial "Johannine" vocabulary. Lk. has 62 favorites, Mt. 37, and Mk. only 18 (so Morgenthaler, 18 if). Thus the 15

"Johannine" words in the longer text (or the 16 in GS) are a fifth of the 75; the 8

"Lucan" words are about an eighth of the 62 (the 6 in GS are a tenth); the 7 "Matthaean" (7 also in GS) are about a fifth of the 37, and the 4 "Markan" (5 in CS) are about a quarter of the 18. Here again, the similarity between the figures for the longer text and those for CS is obvious; and the longer text, like CS, contains a slightly higher percentage of the list of Mk.'s favorite words than it does of the list of any other Gospel's. This, however, may be no more significant than the other data concerning these frequencies.

Morgenthaler's 18 Markan favorites occur 444 times in Mk., that is, about once every 25.22 words. Of the longer text we have just 175 words (181 less the final inter polation και ούκ άπεδεξα το αύτάς ό ' Ιησούς, on which see the commentary). An average

175 words of canonical Mk. should contain 6 or 7 uses of Mk.'s favorite terms. The 175 words of the longer text contain 6 (2 of εύθυς, i of ΐνα, i of κρατεω, and 2 of όπου) ; the 1 75 words of CS contain 5 (one each of άλλά, δύναμαι, δώδεκα, πάλιν, and πολύς). This, again, is perhaps insignificant, not only because of the banal character of most of the "favorites," but also because so small a sample might be expected to diverge widely from the average. Consequently, its agreement with the average may also be mere chance.

Parker, Gospel, has made a careful study of the vocabulary characteristic of both

Mk. and Mt. and has condensed the results into four tables (pp. 41, 245-250) listing with considerable duplication some 1 19 expressions. Of these, 1 1 appear ^n the longer

text [άπερχομ at, εκεΐθεν, εκτείνω, επιτιμάω, εύθύς/-εως, ' Ιορδάνης , κρατεω, οφ ία, παρακαλεω, πέραν, and υίε Δαβίδ), while only 8 appear in GS (ά γρός, άρχιερεύς, άρχω,

THE SECRET GOSPEL

γραμματέας, εμπαίζω, εν rfj όδώ, εύαγγέλι ον, and προάγω). Thus the longer text, with only a sixty-fourth of Mk.'s 11,200 words, contains about a tenth of the MatthaeanMarkan expressions in Parker's list, while CS contains only a fifteenth. This may possibly be significant in view of the evidence seen above (on προσεκύνησε, in II. 24) for supposing that Mt. knew the longer text. Again, however, the commonplace character of most of the words and the brevity of the quotations from the longer text make it impossible to build on these data.

The studies of Hawkins [Horae) and Turner [Usage) have been most valuable because they take into account not only relative frequencies of usage, but also the ways in which words were used — questions of syntax and style. For the same reason, however, their data are more difficult to classify, since some belong properly in the sections on phraseology and grammar. Nevertheless, since Hawkins' classification of his material was primarily lexical, his results may be stated here. He listed 95 "words and phrases characteristic of" Mt. (pp. 4-8). Of these, 5 [γυμνός, εΐς for τι?, εκεΐθεν in narrative, προσέρχομαι, and προσκυνάω) appear in the longer text of Mk. (none appears in GS). Among these, Hawkins thought προσέρχομαι a "most distinctive and important" trait (p. 7 — he did not explain how it happened to occur 10 times in Lk.) and γυμνός "less important than the rest, because mainly or entirely accounted for by the subject matter" (p. 4). For Mk. he listed 41 characteristics, of which 3

[ερχεται/ -ονται historic present, εύθυς, and κρατεω) appear in the longer text, 4 in CS (eV rfj όδώ, εύαγγελι ον, θαμβεω, and πάλιν). He thought εύθυς and ερχεται "most distinctive," κρατεω "less important." Five is one-nineteenth of 95, 3 one-fourteenth

(and 4, one-tenth) of 41 ; so these figures show CS markedly, and the longer text slightly, more Markan than Matthaean. The 41 Markan characteristics appeared in Mk. 357 times, about once every 31.5 words. In both the longer text and CS they should therefore have appeared between 5 and 6 times. They actually appear 5 times in the longer text (2 έρχομαι in historical present, 2 εύθυς, ι κρατεω), 4 in CS (one each of the traits listed above). Hawkins listed 151 characteristics of Lk. (pp. 16-23) ; of these only 3 (redundant άναστάς, πλούσιος, and συν) appear in the longer text of Mk. and only 1 [νυν) in CS. This seems to go beyond chance and to indicate clearly that, like CS, the longer text has little or no connection with Lk. Besides his lists,

Hawkins devoted a chapter to minor Markan peculiarities. One of these (p. 119) is the fact that οργή is attributed to Jesus nowhere in the Gospels save Mk. 3.5 and the western reading in 1.4 1. The latter now seems to have been derived from the longer text. Another peculiarity Hawkins noted was Mark's preference for καί as against δε at the beginnings of sentences. It was shown above (on ό δε, in III. 4) that the proportion of καί to δε in the longer text is almost the same as in canonical Mk.

Turner, Usage, concerned himself chiefly with grammar, but included a number of lexical observations; he noted Mark's fondness for numerals (26.337), his use of ηρξατοβντο as a substitute for the imperfect (28.352), his preference for εκ as against από (Matthew and Luke prefer the latter, 29.281), his use of εκ after εξέρχομαι [ibid.), his fondness for diminutives (29.34911 — cf. νεανίσκος) , and his fondness for iva and "improper" use of it after παρακαλεω (29.356). All these are found in the longer text of Mk.

THE SECRET GOSPEL

To these observations a few may be added, αγαπάω , of Jesus' loving another man, is found in the synoptics only in Mk. άποκνλίω is a rather rare word which Matthew and Luke seem to have derived from Mk. Βηθ ανία, as noted in the commentary, was more prominent in Mk. than in the later synoptics. Mk. uses έμβλέπω twice as often as any other evangelist, and 3 out of 4 times in the έμβλέφας αυτωβοΐς construction used in the longer text, μυστήριον is another word which Mt. and Lk. use only when they parallel Mk. ; in the longer text the use of the singular is specifically Markan.

Σαλώμη is peculiar to Mk., the later synoptists chose to omit her name, σινδών is 4 times as frequent in Mk. as in Mt. or Lk. (Moulton-Geden omits the first instance in 15.46). Whenever Mark mentions the word he repeats it; in any other author this would be thought emphatic, χειρ as an instrument of supernatural help is typical of Mk. : 10 or 11 times, 7 in Mt. (5 Markan), 4 in Lk. (2 Markan).

Combining the above observations with the lists of Morgenthaler, Parker, Hawkins, and Turner, we have:

*Ίορδάν ης αγαπάω (of Jesus) καί (initial, vs. δέ) *κρατέω * άπέρχο μαι άποκνλίω άρχω (for imperfect) μυστήριον νεανίσκος ( qua diminutive) οπού Βηθ αν ία *οφία εκ ( VS . άπο) *εκε ΐθεν οργίζω (of Jesus) 'i /

*€KT€LVOJ

*παρακαλέω *πέραν (το) Σαλώμη έμβλέφας (α υτω) έξ ( qua numeral) εξέρχομαι εκ σινδών *υίέ Δαβίδ * έπιτιμάω έρχεται/ -οντ αι (historic present) ενθνς χειρ (with supernatural power) ΐνα * Listed by Parker as characteristic of both Mk. and Mt.

In all, 29 of the 82 words listed can claim to be, at least in one construction or an other, characteristically Markan. Here it may be objected that to consider special constructions confuses the data on vocabulary with those on phraseology and gram mar, and has resulted in some overlapping of the above list with those in the preceding and following sections. The objection is justified. However, it seemed impossible to exclude consideration of special constructions from an account of the vocabulary, since vocabularies differ not only by the words used, but also by the special meanings given them. If this be granted, it follows that the vocabulary of the longer text is

preponderantly Markan.

Indeed, it is so preponderantly Markan that it must be explained as the result either of the same stream of tradition which produced Mk., or of deliberate imitation

THE SECRET GOSPEL

6/7t b. Phraseology * Mk.

6/7 Mt.

Lk.

— καί + participle of έρχομαι + verb

3/4 — — οι be μαθηταί initial + verb 0770 υ ην

Jn.

— — καί ευθύς initial — — —

2/3 1/2 7/8

I I — i(?) — 4/6 ηρξατο/ -αντο -finfinitive

6/7 1/2

— καί ηρξατο/-αντο initial μετ αύτοΰ, of Jesus' followers

4/5 — ην/ησαν γάρ explicative (καί) μετά . . . ημέρας + numeral

— — — I I / I 2

— —

έκεΐθεν + participle, initial 3/5

I — (In Mt. the participle always precedes, in Mk. and in the longer text it follows.) το πέραν

I —

I I 3/4 — η μητηρ αύτοΰ * For details see Appendix E. f 6/7 means "6 or 7," and so hereafter. The uncertainty reflects textual variants.

J In a different sense.

§ The indented expressions in this list are not found in the longer text of Mk.

THE SECRET GOSPEL

of Mk. (Composition as a cento seems unlikely; the parallels come from too many places, are combined and modified too freely, and fit together too well.) If imitation, it was imitation of a very simple sort. The imitator must have known Mk. almost by heart and deliberately told his story as much as possible in the words and phrases of the original. Therefore we should expect his work to show almost no other words or phrases. On the other hand, if the longer text were a free product of the same stream of tradition, we should expect it to differ from canonical Mk. at least as far as canonical

Mk., in its various parts, differs from itself. Now the quotations from the longer text contain only 3 words not in Mk., and of these three αποδέχομαι was picked out in the commentary on III. 16 as part of a later addition, κήπος and οργίζω remain. But Mk. uses some 634 words once only (Morgenthaler, 166)— approximately one word in every 17.8. The quotations from the longer text (excluding the addition containing αποδέχομαι) have 1 75 words and should therefore have about 10 not found in Mk.

That they actually have only 2 non-Markan words suggests that they were produced by imitation of canonical Mk., not by independent composition. But this evidence, again, is not conclusive: GS has only 4 words not found elsewhere in Mk. The per centile originality of Mk.'s vocabulary would probably decline as the amount of material increased; the contentual similarity of the main story in the longer text to the other stories of cures and resurrections would make for the use of many of the same words; and, above all, the distribution of hapax legomena is uneven and the quotations from the longer text are so short that their variation from the average might be mere chance. Compare Jn., of which Parker states that "the number of words occurring in only one chapter . . . ranges from 2 in chap. 17 to 47 in chap. 19"

( Two Editions 306 ng).

The evidence yielded by this list is clear, especially since Mk. contains only about 11,200 words, Mt. about 18,300, Lk. about 19,400 (Morgenthaler, 164), and the figures for Mt. and Lk. should therefore normally exceed those for Mk. by 7/1 1 and 8/1 1 respectively. In the above list, of the 33 entries which represent usages found in the longer text of Mk. there are 18 in which the number of parallels from canonical

Mk. is greater than that from either Mt. or Lk., and 9 more in which it runs so close to the largest number from either Mt. or Lk. that the difference is insignificant: οι δέ μαθηταί, δπου ην, έκ τον μνημείου , φωνή μεγάλη, έκτείνειν την χ€Ϊρα, και έξελθόντες, όφίας γενομένης, έκεΐθεν + participle, and πέραν τοΰ Ίορδάνον.

Of the 6 remaining, 3 are most frequent in Lk. : μένειν συν, η βασιλεία τοΰ θεοΰ, and η μητηρ αυτόν. The differences in frequency of these are determined by content, not style, and are insignificant for the question of the written sources of the longer text of Mk. — which should not be supposed to have derived its references to η βασιλεία τοΰ θεοΰ or η μητηρ αύτοΰ from Lk. simply because Lk. uses these phrases most often. μένειν συν, too, is ordinary Greek and need not be thought to have been derived from Lk.

The list clearly shows the distance of Lk. from the style of the longer text of Mk. Of the 33 items, 1 1 are not represented at all in Lk., 6 others appear in Lk. less often than in either Mk. or Mt., and for 2 others Lk. is tied with the lower of its competitors.

Mt. is much closer than Lk. to the longer text of Mk. (as it is to canonical Mk.,

Parker, Gospel 32-43). And the traits to which most parallels appear in Mt. are usually

THE SECRET GOSPEL

traits of style, not content (/cat + participle of έρχομαι + verb, /cat προσελθών initial,

/cat εζελθόντες initial, όφίας γενομενης initial, εκεΐθεν + participle, initial). To a num ber of these data Parker called attention in his preliminary report to the Society of Biblical Literature and Exegesis at its New York meeting in December, i960 — an unpublished report to which this study is often indebted. However, in comparison with the full list, the Matthaean traits are insignificant. Mt. leads in only 9 out of 33 instances. In 6 of these 9 (ot δε μαθηταί, εκτείνειν την χοΐρα, καί εξελ θόντες, όφίας γενομενης, εκεΐθεν -f participle, πέραν του Τορδάνου) there are so many Markan in stances that the trait cannot be considered typically Matthaean. In two, indeed, there are minor peculiarities which distinguish the Markan from the Matthaean usage, and the longer text has Markan form (/cat όφίας, not όφίας δε; εκεΐθεν before the par ticiple). Therefore, as truly Matthaean traits we have only the sentence structure

/cat + participle of έρχομαι + verb, έρχομαι followed by προσκυνεΐν, and initial /cat προσελθών. Of these, however, canonical Mk. uses the first 5 times and the last 1 or 2, so they are not alien to Markan style. The only Matthaean trait found in the longer text, but not in canonical Mk., is the use of έρχομαι to introduce προσκυνεΐν.

Since Mk. 5 times uses έρχομαι to introduce other verbs, and once uses προσκυνεΐν of a petitioner coming to Jesus, it is not improbable that a Markan text should have combined these constructions. That the longer text is Markan rather than Matthaean even in this detail is suggested by the fact that in it προσκυνεΐν governs the accusative, as it does in Mk. 5.6, but never in Mt.'s 13 uses except once, when he is quoting the OT. This argument is not conclusive, because the accusative in the longer text, and also in Mk. 5.6, may be corrupt. However, against the one Matthaean trait of έρχομαι + προσκυνεΐν must be set the list of non-Matthaean traits ( και έρχονται εις, και ευθύς, καί είσελθών, εισέρχομαι όπου, κρατησας της χειρός, ηρζατο + inf., καί ηρζατο initial, οι μετ' αύτοΰ of Jesus' followers, βασιλεία του θεού), most of which are typically

Markan. In Hawkins' list of expressions characteristic of Mt. ( Horae 4ff) there are 27 of which one example each appears in canonical Mk.; it should be expected, therefore, that a longer text of Mk. would contain additional examples of such isolated Markan usages of expressions common in Mt.

The letter's occasional contacts with John are clearly insignificant for the question of its style (as opposed to content), εκ του μνημείου and η μητηρ αύτοΰ are determined

By content and have no stylistic peculiarity; όπου rjv, while more frequent in Jn., is also found in canonical Mk. and is ordinary Greek. Those who would see in such occasional parallels proofs of dependence should look at section IV of Appendix G, where it appears that πέραν της θαλάσσης of Mk. 5.1 has 3 parallels in Jn., none else where; so does ύπηντησεν αύτω of Mk. 5.2; ούδείς εδύνατο in Mk. 5.3 has 5 parallels in Jn., i in Mt., and none elsewhere; τί εμοί καί σοί in Mk. 5.7 has a verbatim parallel only in Jn. 2.4. Yet no one would take these as proof that Mark used Jn. or John,

Mk. See also the Johannine parallels to the material from Mt. 9, in section V of Appendix G, and the remarks above in the commentary on III. 15.

Finally, it has not seemed worthwhile to undertake a detailed comparison between the quotations from the longer text and the "apocryphal" Gospels. Of the latter,

Thomas, being in Coptic, does not admit of close verbal comparison. The earliest

THE SECRET GOSPEL

material in Greek — the fragments of Thomas , the pericope De adultera , P. Egerton 2, the Gospel of Peter , etc. — are obviously much more remote from canonical Mk. than is the longer text. Therefore close comparison of their vocabularies and phraseology is unnecessary. c. Grammar Many of the grammatical peculiarities of the text have already been mentioned: under phraseology, several types of sentence structure; under vocabulary, the use of ήρξατο with the present infinitive as a substitute for the imperfect, the pleonasms

(εξέρχομαι εκ, cf. άποκυλίω από ), the use of the historical present ( ερχεται , λεγει), of Iva with the subjunctive after παρακαλεω, and of initial καί. To these can be added the use of the accusative after π ροσκυνεΐν (unless it be a later corruption of the dative used by the other evangelists— see the commentary, II. 24). All these have been noted as Markan traits.

In his preliminary report, Parker used the relative rarity of historical presents as an argument against assigning the longer text to the Markan tradition. Disregarding verbs in subordinate clauses and quotations, the text has 3 historical presents (all in formulas, καί έρχονται, καί λεγει αντω, ερχεται), 3 imperfects, and 16 aorists. Again disregarding verbs in subordinate clauses and quotations, the story of the rich young ruler, with which the longer text is closely linked by location and content, has 3 historical presents (all λεγει), 3 imperfects, and 7 aorists; the prophecy of the passion, which followed it and immediately preceded the first quotation from the longer text, has no present, 4 imperfects and 1 aorist; the story of the sons of Zebedee, which stood between the two quotations, has 2 presents ( προσπορεύονται and λεγει) and 6 aorists; Mk. 10.46a, which introduced the second quotation, has a historical present

( καί έρχονται) ; the story of Bartimaeus, which followed the second quotation, has 1 historical present (φωνοΰσιν) , \ imperfects, and 7 aorists. It appears that in this section of his Gospel Mark was using historical presents almost exclusively in formulas ( λεγει , έρχονται, etc.) and using the imperfect and aorist "correctly" in the body of his narratives. That is just what we find in the quotations from the longer text and is also characteristic of most passages of Mk. ; see the study of Zerwick, U liter suchungen

49 ff. It may be, however, that the small number of historical presents argues against the supposition that the longer text was a deliberate imitation of Mk., for the fre quency of historical presents in many sections of Mk. is just the sort of thing — at once obvious and easy — which an imitator would affect. That in this respect the quotations should accord with the sections of Mk. adjacent to them, and not with the popular notion of Mk.'s style, suggests they were not imitations but products of the same tradition.

As other grammatical peculiarites of the longer text may be mentioned three Senfitisms: the use of μία for τις in II. 23, found in Mk., but most frequent in Mt. and

reckoned by Hawkins ( Horae 5 and 30) as a Matthaean trait; redundant αυτός in II.23 and HI. 15, almost evenly distributed through the synoptics; redundant άναστάς in III. 10, which Hawkins (16) considered a Lucan trait, though there are 5 or 6

THE SECRET GOSPEL

instances in Mk. On all these see the commentary, ad locc. This frequency of Semitisms in the longer text might be a sign either of early material or of late imitation —

Semitisms catch the eye and are easy to copy.

Less conspicuous are a number of grammatical details noted by Turner, Usage , as Markan traits: (i) The frequency of parenthetical explanatory clauses, especially of γάρ clauses like those in the longer text, III. 6 ήν γάρ πλούσιος and Ill.qf εδίδασκεν γάρ αυτόν 6 Ίησοϋς, etc. (26.145). (2) The use of the plural, usually of έρχομαι, to denote the movements of Jesus with his disciples and the crowd (26.225). (The later synoptists and MS tradition, concentrating on the Master, regularly replace this with the singular.) There are 3 examples in the longer text, II. 23 και έρχονται, III. 5 και εξελθόντες, 1 1 1.6 ηλθον. (3) The use of the singular, referring to movements of Jesus, immediately followed by a reference to the disciples or the crowd (the reference is usually eliminated by the later tradition— 26.231). Of this the longer text as it stands does not provide an example, but the conclusion of the first quotation, III. 1 1 και άναστάς επεστρεφεν, was immediately followed by και προσπορευονται αντω ' Ιάκωβος και * Ιωάννης ; this is paralleled by Turner's examples from Mk. 1.35^ 2.23; 6.1. (4)

Location of the verb at the end of the sentence (29.352). In the longer text, II. 23 ης 6 ά8ελφος αύτης άπεθανεν. On this Turner's comments: "It is not suggested that these instances are typical of Mark in the sense that this order of words is his normal usage: but they are not inconsiderable in number" (29.355). Kilpatrick, Mission 149^ found that "for Mk. 13, the normal position for the verb is the initial one," but he reports a count of verbs in five pages of Mk. which yielded 40 initial, 66 medial, 24 terminal. In Mk. 10.17-52— -the immediate context of the quotations from the longer text — 52 verbs in independent clauses are preceded by expressed subjects or objects,

18 have no expressed subject or object, and 23 are followed by expressed subjects or objects (these figures do not include the OT quotation in verse 19). Evidently

Mk.'s usage in this respect differs greatly from one pericope to another. See further

Zerwick, Untersuchungen 75ff, whose careful critique of Turner leads him to conclude that in this question it is not possible to determine any characteristically Markan practice.

This concludes our survey of the evidence from vocabulary, phraseology, and grammar. I think it has shown that the longer text is related to canonical Mk. not only by a few conspicuous parallels, but also by a multitude of small details which are more like the details of Mk. than of any other evangelist. The vocabulary shows a distribution of Markan and non-Markan words almost identical with that in an adjacent section of equal length from canonical Mk. More than a third of the words listed are themselves characteristically Markan or appear in characteristically Markan constructions. The phraseology is yet more clearly Markan: 18 items most frequent in Mk., 9 in Mt., 4 in Lk., and 2 in Jn. — and these gross figures require modifications which incline the balance even further to the Markan side. The grammar throughout x34 accords with Markan usage and shows half a dozen typically Markan constructions.

Above all, there is nothing in the text (except the terminal interpolation) which on stylistic grounds requires us to suppose knowledge of any Gospel save Mk. The text

THE SECRET GOSPEL

could not have been written by anyone who was not familiar with the tradition represented in Mk., but someone familiar with the Markan tradition could have written it — so far as the style is concerned — without knowing Mt., Lk., or Jn. This, I think, is as far as the stylistic evidence will take us. Accordingly, we now turn to the major parallels.

3. THE MAJOR PARALLELS TO THE CANONICAL GOSPELS

The major parallels differ from those already discussed either by size, being so long that it is difficult to explain them as chance collections of cliches, or by peculi arity of content, containing some unusual element which requires explanation. Since these distinctions are matters of degree, there will be differences of opinion as to which parallels should be discussed here.

To begin with, there are those parallels to the western text which seemed evidence for its dependence on the longer text of Mk. : καί έρχονται els Βηθ ανίαν (11.23), καί όργισθείς (II. 25)? καί ηρξατο παρακαλεΐν αυτόν ΐνα μετ' αυτού η (III. 4— 5), and possibly εξετεινεν την χ€ΐρα καί ηγειρεν αυτόν κρατησας της χει ρός (III. 4) and επεταξεν αυτω ό ' Ιησούς (III. 7). If these are evidence of the influence of the longer text of Mk., they throw no light on its origin, save to locate it before that of the western text

(in other words, before 150?) and probably in the same area where the western text arose (Egypt?).

Next there are the similar (but reverse) cases where the present text of Clement's quotations seems to have been corrupted by the influence of the texts of the canonical

Gospels: ην γα p πλούσιος in III. 6 is the most likely example, καί προσελθων . . . άπεκυλισε τον λίθον από της θυρας του μνημείου in III. 1—2 may be another, but καί προσελθων is a cliche and the rest of the parallel is so fixed by content (and appears so late and so sparsely in the history of the canonical text) that no confidence as to the origin of the parallelism can be justified. In any event, such corruptions from the texts of the canonical Gospels are similar to those found in Clement's quotations from canon ical Mk., as shown in Appendix F, and therefore yield no evidence as to the origin of the longer text, into which they were probably introduced by Clement himself or some later copyist (see above, section II. A, end).

Elimination of these leaves the following:

III.5—6 καί μεθ' ημέρας εξ (Mk. 9·2) III. 8 περιβεβλημενος σινδόνα επί γυμνού (Mk. 14.5O

III. ΙΟ το μυστήριον της βασιλείας τού θεού (Mk. 1 1) III. 15 όν ηγάπα αυτόν 6 ' Ιησούς (4 in Jn.)

To these might be added III. 3-4 καί ηγειρεν αυτόν κρατησας της χειρός (Mk. Ι.31) and καί εξελθόντες εκ τού μνημείου (Mk. 16.8), but the parallels to these are only

approximate, that to the former is found only in part of the textual tradition (XBL,

THE SECRET GOSPEL

etc.), and all the elements of both are cliches, except for ηγειρεν αυτόν and του μνημείου, which are determined by content; accordingly these are not of the same class as the six listed above. Of those six, υίε Δ αυίδ ελεησόν με, το μυστήριον της βασιλείας του θεού, and ον ήγάπα αυτόν ό * Ιησούς probably were fixed phrases of early Christian tradition.

(That the third did not come from John is indicated both by the general absence of Johannine traits from the longer text and by the fact that the form in the longer text preserves a Semitism, αυτόν, which the Johannine form has eliminated.) The appear ance of these fixed phrases in the longer text is no more evidence of imitation than it is of originality. But the other three, εμβλεφας αύτω ηγάπησεν αυτόν, και μεθ ' ημέρας εξ, and περιβεβλημενος σινδόνα επί γυμνοΰ, are phrases of Markan narrative and their recurrence requires special explanations.

Explanations are suggested both by Mark's practice of self-repetition and by repetitions of Markan material in the earliest of the expanded texts of Mk. which are still preserved (that is, in Mt. and Lk.). As shown above, Mark frequently repeats individual words, narrative phrases, and basic sentence patterns. He also tells stories so much alike that they are generally thought different accounts of the same event

(the feedings, the prophecies of the passion), but in these different accounts he does not usually duplicate exactly much of the wording. He often tells several stories of the same type and when he does so he is apt to use the same phrases for similar situations

(see above, on εξετεινεν την χ^ΐρα and κρατησας της χειρός). Within individual stories he is fond of repeating phrases or even clauses (2.5,9,11,12,15,16, etc.; note in Ap pendix G, section IV, the parallels to Mk. 5 from Mk. 5 and 6). Finally, in different stories he will repeat striking phrases or entire sayings to indicate some connection between the different events: Mk. r.11 || 9.7 φωνή εγενετο . . . ό υιός μου ό αγαπητός at baptism and transfiguration; 1.24 || 5.7 τι ημΐν καί σοί, ' Ιησού , the demons' con fession; 4-9 II 4·23 II 7· ός εχει ώτα άκουειν, άκουετω, to indicate that something has been withheld; 5.31 II IO.52 V πίστις σου σεσωκεν σε', 5*43 II 8.36 K0LL διεστείλατο αυτοις

. . . ίνα μηδείς/ μηδέν ι; 6.15 || 8.28 Ιωάννης ό βαπτίζων . . . άλλοι . . . ' Ηλείας . . . άλλοι δε

. . . εις των προφητών, 9-2 || 14-33 παραλαμβάνει . . . τον Πέτρον καί τον ' Ιάκωβον καί τον

Ίωάννην (cf. 5-37) ? 9-35 II ΙΟ·44 TL$ θελει πρώτος είναι εσται πάντων . . . διάκονος.

In most of these the MS tradition shows a tendency to eliminate differences between the parallels. A similar tendency may be supposed to have been at work in the trans mission of the letter's quotations from the longer text. [However, as H.K. observes, the longer text was not copied so often as the canonical text and was not subject to the same theological, liturgical, and political pressures. It was therefore less exposed to contamination.] Of these sorts of Markan repetition, the last is closest to what appears in εμβλεφας αύτω ηγάπησεν αυτόν and περιβεβλημενος σινδόνα επί γυμνοΰ. Particularly close to the latter is Mark's repetition of an OT formula in order to identify the Baptist as Elijah without stating the identification directly: Mk. 1.6 ενδεδυμ ενός . . . ζώνην δερματίνην περί την όσφύν αύτοΰ', II Kgs. 1.8 ζώνην δερματίνην περιεζωσμενος περί την όσφύν αύτοΰ.2 If the περιβεβλημενος phrase be a formula connected with baptism (a question to 2. The omission of the Markan phrase by Oit. does not seem to me to warrant the conclusion that it is an interpolation.

THE SECRET GOSPEL

be discussed hereafter, in the section on content), then this method of identifying the two youths by their clothing as baptizandi , without stating the fact directly, is in Mark's manner. So would be the repetition of the temporal specification μεθ' ημέρας εξ, to equate the secret revelation given by the transfiguration with the teaching of the mystery of the kingdom of God (Mk. g.2ff || III. 7-10) ; but here the parallelism may be due to some Sabbatarian interest. Such repetitions both presuppose and expect exegesis of the sort known in the Hellenistic world as συγκρισις προς Ισον and in the midrashim as gezerah shawah (on which see Lieberman, Hellenism 58ff), a technique which takes even minor identites of wording as indications of relations between the content of the passages concerned. Such use of repetition in Mk. is therefore important as an indication that the Gospel was expected to be a text for teaching and that (as Clement said in his letter, 1. 25) some things — in this case, the significance of the repetitions — were deliberately left to be explained by the teacher.

On the other hand, the repetition of εμβλεφας αύτώ ηγάπησεν αυτόν remains some what peculiar, because the action is attributed in Mk. 10.21 to Jesus and in the longer text to the youth in the tomb, whereas, of the above examples of Markan self repetition, all are attributed to the same person except that in 6.15 (to the people) |

8.28 (to the disciples — who, however, are reporting what the people say). Here the relation between Mk. and the longer text is more like that between Mk. and the later synoptics, from which Hawkins made a large collection of such "words differ ently applied" ( Horae 67ff; cf. Dodd, Historical Tradition 331) which he considered evidence for oral, as against written, tradition. Of his examples, the closest is the parallel of Mk. 10.21, where Jesus says to the rich man (whom he loved) εν σε υστερεί, with Mt. 19.20, where the same man asks Jesus rt ert υστερώ; (though here the change is probably due to deliberate correction rather than oral tradition). In Acts we find stories, already told about Jesus in the Gospels, retold about Peter (Acts 9-33ff || Mk.

2-3ff; 9·3βίΓ || Mk. 5.2 iff), and even one of Jesus' miraculous commands put in the mouth of Peter ( ταλιθα κουμι Mk. 5.41 || Ταβειθα άνάστηθι, Acts 9.40). Thus the change of speaker in the repetition of εμβλεφας αύτώ ηγάπησεν αυτόν suggests a relation be tween Mk. and the longer text similar to that between Mk. and Mt. or Acts. On the other hand, the use of ηγάπησεν αυτόν recalls John, for in Jn. Jesus' love of in dividuals and theirs of him is often mentioned; but in the synoptics, although άγαπάω is common, only 4 of its 27 uses refer to specific persons and only 2 of these may refer to Jesus (Mk. 10.21 and perhaps Lk. 7.47). However, neither the parallel to Jn. nor that to the relationship between Mk. and the later synoptics and Acts can be considered decisive. We must leave the problem open for the present, remarking only that, since αγαπάω is used of Jesus' personal relations in the synoptics, the usage can not confidently be attributed to Johannine influence. Moreover, as remarked above, the εμβλεφας formula is a common Markan introduction. How much theory can safely be built on its recurrence with the two words ηγάπησεν αυτόν?

In sum, then, the major parallels between the longer text and canonical Mk. seem mostly due to textual contamination, either of the western text of Mk. by the

influence of the longer text (5 possible instances) or of the longer text by the influence of Lk. and perhaps Mt. (2 possible instances). Of the 6 remaining major parallels,

THE SECRET GOSPEL

4 (υίε Δαβίδ ελεησόν με, τδ μυστήριον της βασιλείας τον θεοΰ, ον ηγάττα αύτδν 6 ' Ιησούς , and περιβεβλη μένος σινδόνα επί γνμνοϋ) are probably phrases fixed in the usage of certain early Christian circles. Their repetition by the original author is no less likely than their use by an imitator; indeed, it can fairly be said that the repetition of such fixed phrases is a Markan trait. On the other hand, μεθ ' ημέρας εξ is probably derived from a recurrent theological pattern and εμβλεφας αντω ηγάπησεν αυτόν remains a problem. Thus the evidence from the major parallels confirms that from the minor stylistic traits : it points almost always to Mk. as the source of the material, shows no strong reason to suppose knowledge of any other Gospel, and leaves the alternatives still open— either a free composition by the same school of tradition which produced canonical Mk., or an early imitation of material now found in the canonical Gospel.

Again, too, the evidence slightly— but not decisively — inclines to the side of early imitation.

4. THE FREQUENCY OF PARALLELS TO THE CANONICAL GOSPELS

It remains to consider the frequency of parallels and their distribution in the mate rial. This is an important question, for one of the commonest results of imitation is a product too much like the genuine article. Thus, among the strongest reasons for thinking the Epistle to the Ephesians a forgery are the facts that it parallels Colossians far more closely and in different places than any undoubtedly Pauline epistle parallels any other. Cognate questions are raised by the other cases of NT parallelism : the relationship of Jude to II Peter, of the transfiguration in II Peter and the last supper in I Cor. to those in the Gospels, and so on. These, together with the synoptics and their parallelism to Jn. have provided evidence for many different hypotheses. The hypothesis of deliberate imitation can appeal to the example of Ephesians and the clumsier Pastorals; Mt. and Lk. show several more or less faithful sorts of copying combined with omission of occasional passages and addition of material apparently from other traditions (though Parker, Gospel , has made a strong case for a close relationship of some of the Matthaean material to the tradition which produced Mk.) ; jn. was variously interpreted as reworking of synoptic material or independent de velopment of cognate traditions until Dodd, Historical Tradition , practically settled the question in favor of independent development. Reworking is seen again in II Peter's use of Jude, and independent development in the various stories of the trans figuration and last supper.

To choose between these hypotheses — and others also possible — it is necessary to determine the frequency, type, and distribution of the parallels between the longer text of Mk. and the canonical Gospels, and to compare these with the frequency, type, and distribution of the parallels between a similar passage of canonical Mk. and the canonical Gospels. Evidence on these points is given in Appendix G.

Unfortunately, Mk. does not afford an exact parallel even for the miracle story

THE SECRET GOSPEL

in the longer text. Markan miracle stories are of several kinds, but none, in point of style, is exactly similar. Miracles told in connection with stories of legal arguments or of sayings are obviously irrelevant here, as are miracles of which Jesus is himself the subject (baptism, transfiguration) and miracles told only generally, in summary accounts of Jesus' work. The rest — the true miracle stories — fall into two groups distinguished by their style. Those of the one group are told verbosely, with much realistic detail, repetition of phrases, and so on; those of the other, briefly, almost in outline, with little more than the necessary data-occasion, parties concerned, trouble, and means of relief (Dibelius, Structure 158-163; Dodd, Appearances 9-10).

Of the first of these two groups, the story of the Gerasene demoniac has been chosen for comparison with the longer text because of the many similarities between the two — similarities pointed out in Parker's preliminary report. Of the second group the story of Peter's wife's mother has been chosen, also because of its parallels: eV . . . e£eX θόντβς ηΧθον ας την οικίαν . . . καί ευθύς . . . καί ττροσεΧθων ηγείρεν αυτήν κρατησας της χαράς.

Comparison of the ways in which these two stories are paralleled by the rest of the material in the Gospels with the ways in which the longer text is so paralleled reveals :

(1) There are more parallels to the longer text than to the canonical texts. (2) Con versely, the longer text has less of the peculiar details which individualize the canonical stories. (3) The elements paralleled in the longer text are usually longer and more significant; many of the parallels to the canonical text are little more than elements of vocabulary (διά τό with infinitive, ύπ αύτοΰ, διά παντός, etc.), while in the longer text whole phrases are paralleled. (4) In the canonical texts the parallels are chiefly to transitional formulas, elements of framework at the beginnings and ends, technical phrases ( ττνεΰμα ακάθαρτον, the formula for exorcism, etc.), and an occasional fixed saying like τι εμ,οί καί σοί, * Ιησού , which evidently circulated in the tradition since variants of it turn up in various stories. In the longer text not only the transitional formulas and framework, but also the main narrative elements are mostly paralleled; the parallels continue throughout the stories, and the number and individual length of the paralleled technical expressions and fixed sayings is conspicuously higher.

These differences must not be exaggerated. Canonical Mk. sometimes does, briefly, parallel itself very closely. Consider, for instance, Mk. 8.22-24: καί φερουσιν αύτω τυφλόν καί παρακαλοΰσιν αυτόν ΐνα αύτοΰ αφηταί καί αττίλα βομενος της χείρος του τυφλοΰ εζηνεγκεν αυτόν εζί ο της κώμης καί τττυσας εις τα ομματα αύτοΰ , εττίθείς τας χεΐρας αύτώ , €7 τηρωτα αυτόν, ε'ί τι βλ εττείς', καί αναβΧεφας εΧεγεν. Here the underlined words are paralleled (solid, verbatim; dotted, approximately) in Mk. 7.32-34, the cure of the deaf mute. Nevertheless, even in these stories there are peculiar details — the men as trees walking, the use of εφφαθα — which individualize the episode, as the cure of Peter's wife's mother is individualized by the specification of the parties concerned. In the more verbose Markan stories, like that of the Gerasene demoniac,

such peculiar material is conspicuous. (Therefore only the first sixteen verses of that story have been presented in Appendix G; they suffice to show the difference.)

THE SECRET GOSPEL

The lifelike details and realistic verbosity of Markan narrative are commonly sup posed to be primitive traits, and primitive they certainly are vis-a-vis Mt. and Lk., who frequently eliminated them. It is not sure, however, that they were primitive in relation to oral tradition. Their realistic details can no longer be taken as evidence of eyewitness tradition and are suspiciously like the developments in later apocryphal

Gospels (Nineham, Eyewitness 22). Dibelius, Structure 161, 165^ has maintained on literary grounds that the full Markan miracle stories are secondary expansions of earlier, brief ones. [In this he may be mistaken, H.K. remarks; it is possible that the full and the schematic stories represent equally old traditions from different circles of narrators.] However, if Dibelius were right, we should expect that the formulas of the earlier, schematic accounts would turn up again and again in the framework of the later expansions. Now, as already noted, it is the framework of the Markan stories for which there is a plethora of parallels, like that found for the longer text. Accord ingly, the multiple parallels to the longer text can be explained in at least two ways: either this is material earlier than canonical Mk., simpler and more schematic — perhaps a text connected with some important ritual and therefore widely echoed; or this is later than canonical Mk. and shows the Markan phraseology still being used, but in formulaic fashion, as we find it especially in Mt. — the fixed formulas are preserved and the story is told almost entirely in these, the individualistic details having disappeared.

Of these two explanations the first is supported by Clement's statement that the longer text was used in "the great mysteries," and also by the evidence that it has frequently influenced the western text. However, the second explanation is supported by Clement's statement that the longer text was produced by expansion of canonical Mk., and also by the analogy of the existing synoptic Gospels which were so produced.

The analogy is particularly close for certain sections of Mt., which show the same sort of formulaic narration, and have the same plethora of parallels, as does the longer text of Mk. Most conspicuous of these Matthaean passages is 9.27-34, pre sented in section V of Appendix G, which shows all the secondary traits pointed out above as characteristic of the longer text: It has almost no individualizing details, the elements paralleled are often of considerable length, the parallels are not limited to transitional elements, but continue throughout, and they contain not a few fixed sayings like eA όησον ημάς, νιος Δαυίδ. Nevertheless, Dodd has made a strong case (in Historical Tradition i7of) for supposing 9.27-31 an original composition, independent of its Markan parallels and based on a variant form of oral tradition. Other passages in Mt. which present similar problems are 12.22-24 and 21.14-16. Richardson points out the sending of the seventy in Lk. 10. 1-20 — a striking example, but one not in cluded in Appendix G because it draws principally on material from the sayings tradition, whereas Clement's quotations from the longer text are entirely narrative

(though he says that the longer text also contained sayings — λόγιά τινα, 1. 25).

Given this similarity of type between Mt. 9.27-34 and the longer text of Mk., it is important to notice that most of the parallels to the Mt. passage are found in Mt., most of the parallels to the longer text are found in Mk. The figures are as follows :

THE SECRET GOSPEL

To Mt. 9.27-34: Mt. 74, Mk. 25, Lk. 21, Jn. 38.

To the longer text:3 Mt. 84, Mk. 143, Lk. 68, Jn. 33.

This evidence is even more impressive when the different sizes of the Gospels are recalled

(Mt. 18,300 words; Mk. 11,200; Lk. 19,400; Jn. 15,400). Given these figures there can be no doubt of the peculiarly close relation of the longer text to the Markan tradition.

The longer text, without the final interpolation (to which, by the way, there are no verbatim Gospel parallels), contains 175 words, for which Appendix G shows 328 parallels. Mt. 9.27-34 contains 112 words, for which there are 158 parallels. Thus the extent of parallelism to the longer text is substantially higher. This may be in significant, since 99 of the parallels to the longer text are afforded by the three phrases καί €υθυς (2j), καί ζλθοϋσ a (22), and της βασιλβίας του θεού (50), whereas Mt. 9.27—34 has only one phrase for which the figure is above the teens (Aeyet αντοΐς 6 Ίησοΰς,

42).

The high frequency of parallels in the longer text affords support for a special theory of imitation which has been suggested independently by P. Benoit and R. Grant, viz. : The longer text is. a cento produced from the texts of the canonical Gospels.

Grant supports this theory by reference to Irenaeus (Harvey, Li. 15—20 = Stieren,

1.8. 1-9.5). Irenaeus is there attacking the Valentinians. He says that, since they have a theory which neither the prophets proclaimed nor the Lord taught nor the apostles handed down, but which they read out of άγραφα (that is, uncanonical works,

Harvey), they try to twist dominical, prophetic, or apostolic sayings to fit their teach ings, so as to have some evidence for what they say, and to this end they neglect the order and context of the scriptural passages they use and also distort them. He com pares their treatment of Scripture to the breaking up of a mosaic in order to make a different picture with the same tessarae. The examples he gives to illustrate this, however, are examples of allegorical or esoteric exegesis of individual sayings or passages of the canonical Scriptures and afford no evidence for the composition of new, pseudo-Scriptural centos. However, he goes on to say (Harvey, I.1.20, middle =

Stieren, I.9.4) : "Then, collecting scattered expressions and terms, they transfer them, as we said, from the (sense they have) in reality to an unreal (sense) much as do those who set themselves any handy themes and then try to treat them in lines from the Homeric poems, so that less experienced readers might think Homer had com posed the verses about the themes treated ex tempore ." This he illustrates by an ex ample of a Homeric cento, excusing himself by saying, "There is no reason not to cite even such verses, since both (the composer of the cento and the Valentinians) are attempting a similar and, indeed, identical feat." And he concludes that, as the man acquainted with Homer will recognize the verses, but not the theme, and by referring the verses to their proper contexts will show the theme to be spurious, thus

3. When alternate numbers of parallels to phrases in the longer text are given in Appendix G, the lower ones have been counted, since the higher reflect readings with weaker MS support and these are not commonly reported by Moulton-Geden, from which the numbers of parallels to the Matthaean passage have been derived for comparison.

THE SECRET GOSPEL

the true Christian "will recognize the terms from the scriptures and the expressions and the parables, but will not recognize this blasphemous theme." He will acknowl edge the tessarae, but not the picture which has been made of them, "and, referring each of the things said to its proper place and fitting it into the body of the truth, he will expose their fiction and show it to be unsubstantial."

On the strength of this passage, Grant has suggested that the longer text may be a gnostic work of the sort attacked by Irenaeus. However, the longer text has no con nection with the Valentinians, and though it was used by the Carpocratians it was also used by Clement's church, which is commonly supposed to have been orthodox.

Clement expressly asserts that the Carpocratians got it from the orthodox (that is, from his church), and nothing in the text is clearly gnostic. Therefore there is no reason to associate the text with the Valentinian centos, unless it can be shown to be a cento, which is the point in question. Further, the text of Irenaeus does not precisely say that the Valentinians made centos. Irenaeus may have intended to give that impression. [But in the opinion of E.B. he actually had in mind compositions like the Qpmran hymns, which are full of OT echoes but are not true centos. He introduced the bit of cento merely to give his Greek readers the best example he could of the sort of thing he had in mind.] At all events he does not explicitly state that the Valentinians made centos, and — what is most important — he does not produce and demolish any Valentinian cento. This suggests that either he had no such document and was merely using the comparison as a reductio ad absurdum of their neglect of context in exegesis, or he had a Valentinian Gospel which paralleled the canonical Gospels in many places and which he wished to discredit, so he charged that it was a cento but did not give an example from it for fear of discrediting his charge.

On the other hand Irenaeus (Harvey 1. 20. 2 = Stieren, 1. 25. 4) quotes a Carpocratian version of the counsel to be reconciled quickly with one's adversary, which alternately parallels Mt. 5.25 and Lk. 12.58 in a way that can be interpreted as deliberate choice of elements suited for Carpocratian exegesis (Grant-Freedman, 95).

And Dodd, New Gospel 24ff, has practically proved that the text on fragment 1 verso of P. Egerton 2 is a cento of Jn. 5.39,45 and 9.29. The cento form goes back in Greek tradition at least to Aristophanes, Pax 1090-1094, and appears in the OT with the psalms in Chronicles and Jonah. So the possibility that the longer text was produced as a cento is undeniable. And there is no necessity of connecting the cento form with the gnostics: Paul used it in Rom. 3.11-18; Tatian's Diatessaron , the most famous example of the form, was not a gnostic work; and even Mt. and Lk. could be con sidered, loosely, as centos compiled from Mk., Q, and other sources. So the question of form and method of composition need not be confused by introducing the question of doctrinal affiliation.

In favor of the cento theory is the high frequency of parallelism and particularly the frequency of the long parallels discussed above. Against it, however, are the following facts: (1) Some elements of the longer text are not paralleled from the canonical Gospels, which would be impossible were it a true cento of the canonical texts. (2) The great majority of the parallels are brief formulas, most of them used

THE SECRET GOSPEL

many times in the canonical Gospels and more likely to have been put together freely by an imitator than to have been picked out laboriously from here and there by the compiler of a cento. (3) The text cannot be made up by drawing elements from only two or three stories; to suppose it a cento, one must also suppose that the author derived his scraps from practically every chapter of Mk., to say nothing of the other

Gospels — not a likely procedure, especially in antiquity, when most writers, even in citing explicitly, cited from memory.4 (4) Many details in the text do not look as if they had been produced by the compiler of a cento; see, for examples, the notes on 7 τροσεκύνησε in II. 24; νεανίσκος , III. 3; κρατησας της χειρός, III. 3-4, and καί εξελ θόντες εκ του μνημείου, III. 6. (5) The text is too well constructed and economical to be a cento: there are no irrelevant details, every word comes naturally in its place, the narration moves without delays or jumps. Possibly some leisured litterateur might have succeeded in piecing together such a text from the phrases available in the canonical Gospels, but the easier explanation is to suppose it a free composition. (6)

The hypothesis that the text is a cento requires the supposition that someone went to great pains to imitate the style of Mk. as closely as possible, since making a cento is the most laborious, but the closest, kind of imitation. But the longer text is datable, by external evidence, before 125 (see above, section I of this chapter) and at this date

Mk.'s prestige was not high enough to motivate this sort of imitation. Matthew and Luke, in their expansions of Mk., made no attempt to imitate his style. Jn. and the earlier apocryphal gospels ( Hebrews , Thomas , Egyptians , Peter) show no considerable effort to imitate synoptic style. [E.B. remarks that almost none of the early apocryphal

Gospels are even attributed to canonical authors.] So to suppose the longer text a cento would be to suppose it a work unparalleled and unlikely in its time. (7) Finally, it is worth recalling the considerations which led Dodd (New Gospel 35ff), after he had proved that one fragment of P. Egerton 2 was a cento, to conclude that the rest was not (a conclusion he has now reasserted more strongly in Historical Tradition 328 n2). He observed that the attempt to explain all variations of early Christian tradi tions as editorial rehandling of written sources had been discredited by form criticism, which had demonstrated the oral prehistory of the written material and the possi bility that variations might have arisen in the oral period. He found evidence of the realization of this possibility in the story of the centurion's servant in Mt. and Lk. and the nobleman's son in Jn. — undoubtedly the same story, yet in forms so different that literary dependence seemed to him unlikely. In the unparalleled elements of P. Egerton 2 he found evidence that the author was using sources other than the canonical Gospels; he therefore concluded that the divergent forms of canonical

4. Compare the remarks of J. Pouilloux in Fondation Hardt, Entretiens sur Γ antiquite classique, X,

Archiloque, Vandoeuvres-Geneve, 1963, 172k on the use of Homeric formulas by Archilochus and the imitation of Homer by Apollonius Rhodius. Apollonius never draws on the whole of the Homeric poems at once, but imitates one passage at a time, whereas the poems of Archilochus are made up almost entirely of Homeric formulas and variations of Homeric formulas, but the formulas come from all parts of the poems and no one passage is imitated. Archilochus was composing freely in the Homeric tradition, as D. Page showed in the paper (

Archilochus and the Oral Tradition , iigff) which Pouilloux was discussing. The longer text is related to Mk. as Archilochus to Homer, not as Apollonius.

THE SECRET GOSPEL

stories which P. Egerton 2 contained were probably also derived from noncanonical sources. And he remarked in conclusion that "as the number of apocryphal Gospel documents increases, it becomes less and less plausible to suppose that they all originated in expansions of material derived from the canonical Gospels" (p. 48).

These arguments Dodd has now, in Historical Tradition , greatly developed and applied at length to the study of the Johannine problem. Much of his reasoning therein is applicable, mutatis mutandis, to the present case.

If, for the reasons just given, we reject the cento theory, we are left with the alter native previously proposed between free imitation and independent composition in style fixed by the Markan tradition— that tradition of which, by this hypothesis, different elements would appear in the stories of both the canonical and the longer texts. Perhaps, however, the alternative between "imitation" and "composition in traditional style" is false. As remarked above, at this early date "imitation" can hardly have been deliberate faking — Mk.'s prestige was not yet so high as to motivate a forger. "Imitation," therefore, would have been a conscious effort to perpetuate the style of the Markan tradition to which the writer was evidently attached, to express in the traditional phraseology the material now being written down, and to attach it to already written stories by the established Markan technique of repeating phrases as cross references. But this would be practically the same thing as "composition in traditional style."

A similar problem is posed by the remains of early Greek oral poetry which, like the synoptics, was largely written in fixed formulas. The relations of poetic compo sitions of this sort and the history of their gradual reduction to writing have been elucidated by Parry ( Studies ), Lord (Singer), and Notopoulos (Homer', Hymns). Parry's formulaic analysis of the beginning of the Iliad is given at the end of Appendix G; its similarity to the preceding analyses of Gospel material is obvious. Notopoulos, Hymns

343-347, is particularly interesting for the question of transition from free develop ment to exact memorization and written preservation.

5. CONCLUSIONS FROM THE STYLISTIC EVIDENCE

With the phenomena of oral literature in mind, we can see the literary problem before us as that of placing the longer text in relation to the other remains of a tradi tion which was only gradually being fixed in writing. For this purpose, let us review the evidence presented by the preceding stylistic study: (1) The longer text contains nothing which Mark could not have written — nothing incompatible with the canonical

Gospel or without analogy there — except for the final clause of the second quotation, a clause which seems a later addition. (2) Its vocabulary is largely neutral — made up of words used by all the evangelists — but insofar as it inclines toward the vocabulary of any of the Gospels it is Markan (29 words in a list of 82). (3) Its phraseology is predominantly Markan (18 items out of a list of 33, in which 9 of the remaining 15 are neutral). (4) Its grammatical peculiarities, in relation to Gospel usage, are few

and mostly Markan. (5) It is connected with the canonical text of Mk. by 6 major parallels, of which 4 are peculiar to Mk. On the contrary, of the major parallels which might seem to connect it with other Gospels, none affords convincing evidence. (6)

THE SECRET GOSPEL

It has far more parallels to Mk. than to any other Gospel. (7) It contains nothing which necessitates a supposition that the author knew any canonical Gospel other than Mk. On the other hand: (8) It uses only 3 non-Markan words, whereas an equally long section of canonical Mk. might be expected to use 10 (though the section actually tested used only 4). (9) It contains more Markan clauses found in other sections of Mk. than would an equally long section of canonical Mk. (10)

Besides these long clauses, it has many more minor parallels to phrases of the canonical

Gospels than would an equal section of canonical Mk., and these parallels are in dividually longer and are distributed more evenly throughout the text than they would be in a section of canonical Mk. (1 1) In these last three characteristics (8, 9, and 10) it resembles Mt., and particularly Mt. 9.27-34. (12) But there are a few small pieces of evidence (commentary on προσ^κυνησεν, 11.24? on KaL ττροσζλθων απεκυ\ισ€ ,

III. i and on καί ησαν €Κ€Ϊ, III. 1 5) which suggest — but do not suffice to prove — that it was known to Mt.

Of the points above, 8-1 1 all are aspects of one essential fact, the plethora of parallelism. This has been said to admit of two explanations — either that the piece was early and widely imitated, or that it was late and highly imitative. Point 11, the similarity to Mt., is prima facie evidence for a late date; point 12, the evidence suggesting it was known to Mt., for an early one. Both possibilities remain open, though the later date has the strong support of the tradition reported by Clement — that the longer text was an expansion — and the analogy to the other synoptics. The longer text would then be an expansion of Mk. by addition of further material from the Markan tradition, as Mt. and Lk. were expansions of Mk. by addition of further material from the ζ) tradition and other traditions accessible to their respective editors.

As for the date of the longer text : there seems to be no stylistic evidence indicating any date later than that of Mt. A date somewhere between canonical Mk. and Mt. is suggested not only by the evidence for Mt.'s use of it, but also by the consideration that an expansion of Mk. with material from the Markan tradition might be expected to have preceded expansions with alien material. But this consideration has little more than rhetorical plausibility to recommend it. A much stronger reason for an early date is the absence of any clear evidence of knowledge of any Gospel save Mk.

Mt. and Lk. seem to have eclipsed Mk. early (to judge from the indices locorum of the apostolic fathers and the apologists), so it is unlikely that an author writing long after their composition should have drawn chiefly on Mk., and it is almost incredible that, had he known Mt. and Lk., he should not have left in his phraseology many unmis takable traces of his knowledge. But the only strong argument for supposing knowl edge of Lk. (apart from the terminal interpolation) is rjv γάρ πλούσιος, and its isolation makes it easier to explain as a corruption of the text than as an original element.

(If original, why isolated?) There is no strong argument for supposing knowledge of Mt. Nor is there any valid stylistic evidence to indicate knowledge of Jn. (and it

will be shown below that the similarities of content do not indicate such knowledge).

Finally, to this stylistic study of the longer text must be added the statement that the above conclusions are by no means conclusive. The quoted fragments are so

THE SECRET GOSPEL

short that it would be foolhardy to take them as fairly representative of the lost mate rial, or to build on the stylistic data which they afford any considerable theory.

Metzger, Reconsideration , and Cadbury, Dilemma , have argued that the evidence afforded by the Pauline corpus is inadequate to settle the problem of the authenticity of Ephesians. A fortiori .... Perhaps even more in point is the case of the pericope adulterae , of which Cadbury's discussion (Case) leads to results so pertinent for the pre sent investigation that I summarize the article here: Lk. "has the most distinctive vocabulary of any New Testament writer, and a style so individual as to be recog nizable in nearly every verse." And in the pericope adulterae "there are a few unques tioned words that are really characteristic of Luke," as άπο του νϋν, άρχομαι από, eVt/xAtu, ehrev be, ώς (= when). And besides these there are a number of Lucan expressions attested by some, but not all, MSS, and of expressions thought to be

Lucan, but perhaps limited to LukeActs by mere accident. Ergo, "it can safely be affirmed that the passage in its oldest form contained as much distinctively Lucan language as the average passage of equal brevity and simplicity in Luke's acknowl edged works." However, the best MSS omit the passage altogether, and when it is found it is almost always located in Jn. 8 or at the end ofjn. The Lerrar group alone places it after Lk. 21.38. Therefore: "Either (1) the pericope adulterae is an original part of Luke's Gospel and was omitted without leaving any appreciable trace in the MS tradition of that Gospel, or (2) it is written by another than the third evangelist in a style that completely matches his own ... If the first solution is the correct one, then we must believe that in spite of their age, multiplicity and agreement, our authorities for the New Testament text do not preclude such radical divergence from the autographs as the complete omission of a considerable section from one of the four Gospels . . . Here, . . . we should have a flagrant case of primitive tampering, for the omission could only be intentional ... If, on the other hand, the passage is not from the pen of the auctor ad Theophilum, then some one . . . wrote a style that is indistinguishable from the most distinctive of New Testament styles. In this case style proves to be a most unreliable criterion, and all critical arguments drawn from identity of style — such as the common authorship of John and I John, of Luke and Acts, of the Pauline letters, and even of the separate parts of a single work — lose some of their weight."

Since stylistic arguments are thus inconclusive, we turn to questions of content.

III. Structural relations to sections of the

CANONICAL GOSPELS

A. Other miracle stories of the same type In discussing the verbal parallels to the longer text we have already mentioned its similarity in content to the stories of the Gerasene demoniac and of Peter's wife's mother. Parker, in his preliminary report, called particular attention to the former and pointed out a series of details in which it resembled the story in the longer text.

THE SECRET GOSPEL

To estimate the significance of these similarities, we must consider the extent to which the stories in the canonical Gospels are also similar to each other. Fortunately, not all of the stories need be considered, since the resurrection reported in the longer text belongs to a readily recognizable class — that of miracles performed in response to intercession.

One of the historical traits of the Gospels is their account of the revelation of divine power in Jesus as spatially limited; later legend may represent it as bursting on distant strangers (the shepherds, the magi) or producing a general resurrection of the deserving dead (Mt. 27.52); but in the stories which approach historicity, miracles happen when the patients are somehow brought to Jesus' attention or touch his per son. Therefore either the patient must bring himself to Jesus' person or attention, or an intercessor must act on his behalf. Accordingly, most of the miracle stories fall into these two classes (the chief exceptions being miracles in which Jesus himself is the patient — the baptism, transfiguration, resurrection, etc.). The class of miracles in response to intercession comprises the following stories and their parallels: Mk. i.3off

(Peter's wife's mother) ; 4-35ff (stilling the storm) ; 5·22ίΤ (Jai'rus' daughter) ; 7-25ff

(the Syrophoenician's daughter) ; 9. 1 4fT (the demoniac boy) ; Lk. 7.2ff (the centurion's slave) ; Jn. 2. iff (the miracle at Cana) ; 4·46ίΤ (the nobleman's son) ; 1 i.iff (Lazarus).

The intercession may be more or less explicit: Jn. 2. iff is a borderline case; also marginal are Lk. 7. 1 8ff; Mk. 6.35ff — contrast 8. iff and Jn. 6.5ff — and Mk. 9. iff.

All these stories necessarily follow a single basic pattern: situation, intercession, response, miracle. The pattern is found elsewhere, too — with Mk. 7-25ff and g.^ff cf. Philostratus, Vita Apollonii I II. 38. It is obviously useful for resurrections of the dead, though not necessary — one corpse came to meet Jesus (Lk. 7.1 iff) as another met Apollonius (IV. 45). Accordingly, among the nine stories listed above there are two resurrections (Jai'rus' daughter, Lazarus) and two hairbreadth escapes (the cen turion's slave, the nobleman's son). Most of these stories begin with the intercessor's coming to Jesus, and in three of them the intercessor is a woman (the Syrophoenician, the miracle at Cana, Lazarus). Often, moreover, the similarities go far beyond these basic structural elements. For instance, consider the parallels between the Markan story of Jai'rus' daughter and the Johannine story of Lazarus:

The patient is not dead, at first, but only sick.

The intercessors arrive and beseech Jesus.

Jesus' coming is delayed; the patient meanwhile dies.

Jesus declares the dead asleep but is misunderstood.

He reassures the relatives and demands that they believe.

The intercessor falls at his feet.

He sees the mourners weeping and is angry or puts them out.

The mourners do not believe he will be able to raise the dead.

He then goes to the body.

He calls the dead by name or title and orders him or her to arise or come forth.

In response to his command the dead arises and walks.

Jesus gives directions for further treatment.

THE SECRET GOSPEL

Of all these parallels between the Lazarus story and that of Jai'rus' daughter, only the ones underlined are also found clearly stated in the longer text of Mk. By way of contrast with these it is worthwhile to list the similarities of the resurrection story in the longer text of Mk. to the stories of the Gerasene demoniac and Jairus' daughter:

Gerasene demoniac J aims' daughter intercessor Jesus arrives falls at his feet is met man from the tomb beseeches him 7 τροσκννβΐν with accusative άπηλθεν μετ ' αντοΰ trouble with disciples φωνή μεγάλη nudity (?), clothing return to house Jesus comes to house is angered ( ?) goes in to corpse παρεκάλει αυτόν κρατησας της χειρός raises the dead ΐνα μετ' αύτοΰ η Neither of these lists of parallels is so full as the list of those between the raising of Jairus' daughter and the Lazarus story. Consequently there is no need to think that the story in the longer text has any closer relation to that of the Gerasene demoniac or of Jairus' daughter than the latter does to the Lazarus story. Since all these stories

(and the others of the same type, listed above) so often parallel both the longer text and each other, it is both unnecessary and unlikely to suppose the longer text modeled on any one of them. They are all examples of a familiar type of ancient miracle story and their similarities of content and structure (and sometimes even of phrasing) are to be explained as consequences of their common type, not as traces of literary dependence.

B. The Lazarus story However, within this type, the resurrection story in the longer text is particularly close to that of Lazarus. Admittedly there are important differences between them, as Parker pointed out in his report: one sister instead of two, nameless characters, the cry from the tomb, Jesus' rolling away the stone and himself raising the youth.

But similar differences are to be found, for instance, even between such synoptic parallels as the healing of blind Bartimaeus in Mk. 10.46 and the healings of two nameless blind men in Mt. 9-27ff and 20.29ff. And besides synoptic parallels, the Gospels are remarkable for the frequency with which the stories they contain seem to be different versions of the same story. First there are the unmistakable Johannine parallels to synoptic accounts — the cleansing of the temple, the feeding of the mul titude, the walking on the waves, the anointing and the passion story. Whether these result from literary dependence or from common tradition is a matter of well-known

THE SECRET GOSPEL

and unending dispute, but literary dependence has come to seem the less likely expla nation (Haenchen, Probleme, and now Dodd, Historical Tradition). Next, there are many more remote, but unmistakable, parallels which can best be explained as divergent forms of the same tradition: the call of the first four disciples in Lk. and that in Mk.-Mt. ; the miraculous draft of fishes in Lk. and in Jn. ; the rejection at Nazareth in Lk. and in Mk.-Mt.; the centurion's slave in Mt.-Lk. and the nobleman's son in Jn.; the anointing in Lk. and in Mk.-Mt.; Peter's confession in Jn. and the synoptics; the parables of the pounds and talents in Mt. and Lk. ; the feedings of the multitude in Mk. (where two different versions are found in a single Gospel) ; the sendings of the twelve and the seventy (here the two different versions are both in Lk.) ; the gift of the power to bind and loose, in Jn. and Mt. (here two versions in Mt.) ; the demand for a sign and the Beelzebub charge (also two versions in Mt.) ; the passion stories in Lk. and in Mk.-Mt. How far this variation may go, it is hard to say. Richardson, for instance, thinks the ten lepers of Lk. 1 7.1 iff a gentile development of that story of which an earlier version appears in Mk. i.4off. The transfiguration has often been thought a version of some resurrection story (recently by Garlston, Transfiguration , but cf. Burkill, Revelation i6of and ni7; Dodd, Appearances 25 and Close , passim ; Bultmann, Geschichte 65; etc.; an interesting classification of the material is found in Stromsholm,· Examination 255f). Even more — -and yet more remote — examples will be found in Dodd, Historical Tradition, especially in part II, ch. 1, pp. 315-334.

We shall come back later to the question of more remote relations, when we con sider the parallels between the Lazarus story and the stories of Jesus' resurrection.

Here it is enough to have established that this sort of relation is typical of Gospel material and is found within each of the synoptics, and between any two of the synoptics, and between each of the synoptics and Jn. Indeed, related stories are so numerous that a more cautious and convincing type of form criticism might have resulted from the study, not of types of stories, but of these cases, in which comparison might make it possible to determine rather precisely the developments. See, for ex ample, how the following analysis of the Johannine Lazarus story, as indicated by the longer text, differs from the analyses proposed by Bultmann, Johannes, ad loc., and by Wilkens, Erweckung.

Comparison of the Lazarus story in Jn. with that in the longer text of Mk. must begin with the observation already made by Parker, that both occur at the same place in Jesus' career. Jesus has gone up from Galilee to Judea, and thence to Trans jordan. Therefore it is worthwhile to begin with the parallel between Jn. 10.40 and canonical Mk. 10.1 and go on from this to the Lazarus story:

MARK

JOHN 10. 40-ΙΙ. 54

40 ELai άπηλθεν πάλιν πέραν του Τορδάνου εις 10. i καί εκεΐθεν άναστάς έρχεται els τά όρια της τον τόπον όπου ην ' Ιωάννης τό πρώτον βαπτίζων,

41 καί εμενεν εκεί, καί πολλοί ηλθον προς αυτόν '/ουδ αίας καί πέραν του Τορδάνου, καί συνπορευοντα ι πάλιν οχλοι προς αυτόν, καί ώς ειώθει πάλιν έδιδα σκεν αυτους.

καί ελεγον ότι Τωάννης μεν σημεΐον εποίησεν ούδε'ν, πάντα δε όσα εΐπεν Τωάννης περί τουτου

THE SECRET GOSPEL

αληθή ήν. και πολλοί ε'πιστευσαν els αύτον ε'κεί 2-12 dispute on divorce 13-16 blessing children 1 7-23 the rich young ruler 24-31 sayings on rewards

*Ην Se τις ασθενών, Λάζαρος από Βηθαν las έκ τής κώμης Μαρίας και Μάρθας τής άΒελφή

αύτής. ήν δε Μαριαμ ή αλ€ΐφασα τον Κύριο μνρω και έκμάξασα τούς πόδα? αύτού τα ΐς θριζί

αύτής, ής 6 άά€λφος Λάζαρος ήσθέν€ΐ. άπεάχτει λαν ονν αί άδελ</>αί προς αύτον λέγουσαι, Kvpu

toe ον (piAeis a auevet. ακουσας oe ο Ιησού είπεν, Λύτη ή άσθέν€ΐα ούκ εστιν προς θάνατο άλλ' υπέρ τής Βόζης του Θεοσ, ΐνα Βοζασθή

i ιος του Oyeov οι αυτής, ήγαπα oe ο Ιησούς τή Μάρθαν και την αδελφήν' αύτής και τον Λάζαροι ώς οΰν ήκουσ€ν οτι άσ#ενει, τότε μεν έμεινεν έ

ώ ήν τόπω Svo ημέρας· έπειτα μετά τούτο λέγι 32 ήσαν δε έν τή όδώ άναβα ίνοντε? els ' /εροσό - J λύμα, καί ήν προάγων αυτούς 6 ' Ιησούς , τοΐς μαθηταΐς, ' Λγωμον els την Λ ο vS αίαν πάλιι λέγουσιν αύτω οι μαθηταί, Ραββ€ΐ, νυν έζήτον καί έθα μβούντο, οι Se άκολουθούντ€ς έφο8 βοΰντο. καί παραλαβών πάλιν τούς δώδεκα oe λιθάσαι οι '/ουδαίοι, και πάλιν ύπάγ€ΐς ε'κεί απεκρισρ ιησούς, υυχι οωοεκα ωραι ei σιν τη ήρζατο αύτοΐς λέγ€ΐν τα μέλλοντα αύτω 9 ημέρας ; έάν τις ποριπατή έν τή ημέρα, ού προσ

33 συμβαίν€ΐν, οτι '/δου άναβαί 'νομ€ν els 'Iepoσόλυμα , και 6 νιος τον ανθρώπου παραδοκόπτει, ότι το φως τού κόσμου τούτου βλέπ€ΐ έάν 8έ τις περ ιπατή έν τή νυκτί, προσκόπτει, οι θήσ€ται τοΐς άρχ ιερεάσιν και τοΐς γpaμμaτeύI Ο

σιν , και κατακρινοΰσιν αύτον θανάτω και το φως ούκ έστιν έν αύτω. παραδώσουσιν αύτον τοΐς eOveaiv και έμ34 παίξουσιν αύτω και έμπτύσουσιν αύτω και ΐ4

τούτο λέγ€ΐ αύτοΐς, Λάζαρος 6 φίλος ημών kckoi καί μετά τρεί? ημέρας άναστ^σεται μηται· άλλα πορεύομαι Γνα έζυπνίσω αύτόι είπαν οΰν οι μαθηταί αύτω, Κύρΐ€, el κ€Κθίμηται σωθήσ€ται. είρτ^κει δε ό ' Ιησούς περί τού θανά του αύτού· ε'κεινοι δε έ'δο^αν ότι περί τής κοιμή

' Ιησούς παρρησία, Λάζαρος άπε^ανεν, / κο χαίρω Si υμάς, ΐνα πιστεόσρτε, ότι ούκ ή μη ι6 ε'κεί· άλλα άγωμ€ν προς αύτον. είπεν οΰν Θωμά

6 λογόμ€νος ΔίΒυμος τοΐς συνμαθηταΐς, ' Αγωμ€ 42 εκεί] om ρ45 ιι8 pc lat sy ac2 bopc 11. 2 δ« add αυτή X e: add αυτή η ρ45 6 ev ω ην ] εττι τω ρ45Ι

7 επειτα] ειτα p45vid,66p) | τοι$. om p45vid,66* e 1 1 κ€κοψηται ] κοιμάται D lat, item vs. 12 D latt

THE SECRET GOSPEL

THE LONGER TEXT

καί ημείς ινα άποθάνωμεν μετ αύτοΰ. Έλ11.23 καί έρχονται ι8 θών οΰν δ 'Ιησούς εύρεν αύτδν τέσσαρας ήδη ημέρας εχοντα εν τω μνημείου, ήν δε Βηθ ανία εις Βηθ ανίαν καί ήν εκεί εγγύς των 'Ιεροσολύμων ώς άπδ σταδίων δεκα πέντε. πολλοί δε εκ των ' Ιουδαίων έληλύθεισαν προς την Μάρθαν καί Μαριαμ, Ινα παραμυθήσωνται αύτάς περί του αδελφού, ή οΰν Μάρθα

.24 άπέθανεν aneuavev ο αοβλφος μου. και νυν ο coa on οσα αν αίτηση τον Θεόν δώσει σοι ό Θεός, λέγει αυτή

ό ' Ιησούς , Άναστήσεται ό αδελφός σου. λέγει αύτω ή Μάρθα, Οίδα ότι άναστήσεται εν τή άναστάσει εν τή έσχάτή ημέρα, είπεν αυτή ό ' Ιησούς ,

Έγώ είμι ή άνάστασις καί ή ζωή' ό πιστεύων εις εμέ καν άποθάνη ζήσεται, καί πας ό ζών καί

πιστεύων εις εμέ ου μη άποθάνη εις τον αιώνα' πιστεύεις τούτο; λέγει αΰτω, Ναί, Κύριε· εγώ πεπιστευκα οτι συ ει ο Λριστος ο ϊ ιος του Ιϊ/εου ο εις τον κόσμον ερχόμενος, καί τούτο είποΰσα άπήλθεν καί έφώνησεν Μαριαμ την αδελφήν

αυτής λάθρα είπούσα, 'Ο Διδάσκαλος πάρεστιν καί φωνεΐ σε. εκείνη 8e ώς ήκουσεν, εγείρεται ταχύ καί ήρχετο προς αυτόν οϋπω δέ έληλύθει ό Ιησούς εις την κώμην, άλλ' ήν ετι εν τω τόπου όπου ύπήντησεν αΰτω ή Μάρθα, οί ούν Ιουδαίοι οί όντες μετ αυτής εν τή οικία καί παραμυθούμενοι αυτήν, ίδόντες την Μαριαμ ότι ταχέως άνέστη καί έζήλθεν. ήκολούθησαν αυτή, δόζαντες ότι υπάγει εις το μνημεΐον ινα κλαύση εκεί, ή ούν

καί έλθούσα Μαριαμ ώς ήλθεν όπου ήν ' Ιησούς , ίδούσα αυτόν έπεσεν αυτού προς τούς πόδας, λέγουσα αΰτω, προσεκύνησε τον Ίησούν καί λέγει,

Ιησούς οΰν ώς είδεν αυτήν κλαίουσαν καί τούς έπετίμησαν αυτή, καί .26 όργισθείς 6 Ιησούς συνελθόντας αυτή ' Ιουδαίους κλαίοντας, ένεβριI 7 Ιησούς ] add εις Βηθανιαν D f IJ 33 pm bopc 1 9 την p66 NBW al; R] τας -περί p45A(9 f/ Ϊ13 pm ς: om D (lat)

21 Κύριε ] om B sys 25 και η ζωη] om p45 a Γ sys Cypr Orpt 28 λαθρα (, ς], λαθρα)] σιωπή D lat sy3 29 εγείρεται ρ45,66ΑΘ f i Ϊ13 pm vgcl,w j] ηγερθη NBDW al it vgs Co; R | ηρχετο NBW pc it co; R] ερχεται p45,66AD0 f 1 Ϊ13 pi lat 9

33 ενεβριμ ... εαυτόν] (13. 2l) εταραχθη τω πν. ως εμβριμουμενος p45>66cD(0) ϊι dp sa ac ac2

THE SECRET GOSPEL

34 μησατο τώ πνεύματι καί ετάραξεν εαυτόν , καί 35 Ππεν, Που τεθείκατε αυτόν; λεγουσιν αυτω,

30 Κύριε , ερχου καί ιοε. εοακρυσεν ο Ιησούς, ελε37 Ύον °νν οί Ίουόαΐοί, "Ihe πώς εφίλει αυτόν, τινεε he εξ αυτών είπαν , Ούκ eStWro οΰτος ό άνοίξας τους οφθαλμούς τοΰ τυφλοΰ ποιησαι ΐνα καί άπηλθεν μετ' αυτής els τον κήπον οπού ην

38 οΰτος μη άπ οθόνη ; ' Ιησούς ούν πάλιν εμβριμώμένος εν εαυτώ ερχεται εις το μνημεΐον ην he το μνημεΐον III . I καί ευθύς ηκούσθη εκ του μνημείου φωνή

39 σπηλαιον , καί λίθος επεκειτο επ' αυτω. λεγει ό ' Ιησούς , 'Άρατε τον λίθον, λεγει αυτω η άhελφη τοΰ τετελευτηκότος Μάρθα, Κύριε, ηάη όζει' μεγάληκαί προσελθών 6 Ίησοΰς 40 τεταρταίος γάρ εστιν. λεγει αύτη ό Ίησοΰς, Ούκ είπόν σοι ότι εάν πιστεύσης οφη την άόξαν τοΰ

.2 άπεκύλισε τον λίθον από της θύρας του 41 Θεοΰ; I ήραν οΰν τον λίθον, ό he Ίησοΰς ήρεν τούς οφθαλμούς άνω καί είπεν, Πάτερ, ευχαριστώ μνημείου' καί είσελθών ευθύς οπού ην ό 42 σοι ότι ηκουσάς μου. εγώ he rjheiv οτι πάντοτε

• 3 νεανίσκος εξετεινεν την χεΐρα καί ηγειρεν μου άκούεις · αλλά h ιά τον οχλον τον περιεστώτα

•4 αυτόν' κρατησας της χειρός· ό he νεανίσκος εμβλεφας αυτω ηγάπησεν αυτόν καί ηρξατο εΐπον, ΐνα πιστεύσωσιν ότι συ με άπεστειλ ας.

43 καί ταΰτα ειπών φωνή μεγάλη εκραύγασεν, .5 παρακαλεΐν αυτόν ΐνα μετ' αυτοΰ ή. καί

.6 εξελ θόντες εκ του μνημείου ηλθον εις την 44 Λάζαρε, όεΰρο εξω. εξηλθεν ό τεθνηκώς heheμενος τούς πόόας καί τάς χεΐρας κειρίαις, καί η οικίαν του νεανίσκου· ην γάρ πλούσιος. όφις αυτοΰ σουόαρίω περιεόεόετο. λεγει αύτοΐς ό

.η-ιο the nocturnal initiation 45-53 the Jews' reactions and plot.

54 '0 ούν Ίησοΰς ούκετι παρρησία περιεπάτει εν τοΐς Ίουόαίοις, άλλά άπηλθεν εκεΐθεν εις την

.10 εκεΐθεν he άναστας χώραν εγγύς της έρημου, εις Έφράϊμ λ εγομενην .1 1 €7 τεστρεφεν εις το πέραν τοΰ ' Ιορόάνου . πόλιν, κάκει εμεινεν μετά τών μαθητών.

39 V αδ. τ. TereA.] trsp post Μαρθα D vg j^p,h sa bo: om Θ it sys ac2 41 λίθον p66XBDW α/lat; R] add ου ην A f / 57p al f: add ου ην ο τεθνηκώς κείμενος f 13 pm ς\ add οπού εκειτο η6 44 τεριεόεόετο ] εδεδετο ρ45

54 εκεΐθεν] om P45D 579 Pc lat Sys ac2 | χωράν ] add Σαμφουρειν D | εμεινεν KBW pc, R] διετριβεν p45,66AD@ f 13 pm latt co s Comparison of these two texts shows that the story in the longer text of Mk. is of more primitive form than that in Jn. The majority of the contentual differences between the two are the results of Johannine5 additions, to wit:

Jn. 1 1. 1-2 : The preface, naming the hero, relating him to Mary and Martha of the same village, and identifying Mary as the woman who performed the anointing in the same village. The basis of this was probably the common name of the village —

Bethany. The naming of unknown characters and the attempt to relate the characters

5. "Johannine" here means "in style and/or content typical of the present Gospel according to John."

!52

THE SECRET GOSPEL

of stories located in the same place are well-known secondary traits (Bultmann, Geschichte joff, and note in Erganzungsheft to p. 72; also 256^ 338, and Johannes

301 n4 end, 302 ni ; etc.; Barrett, 324 on verse 1 end; Bauer, Leben 5i6f). That the naming of Lazarus is secondary even in the Johannine story is persuasively argued by Eckhardt, Tod 22ff. The doubling of the sister is paralleled by the doubling of the blind men in the Matthaean retellings of Mk.'s Bartimaeus story, Lk.'s doubling of the angel in the resurrection, etc. (more examples in Bultmann, Geschichte 345; see further below, on 11.28-31).

1 1.3: The sisters send word to Jesus. This may be from the story which was known to John, but is probably an attempt to provide motivation for Jesus' return to Jeru salem. John was in the habit of inventing historical explanations: 4.1,45; 5.16,18;

6.2, i4f, 22ff; 7.1,5,30; 8.20; n.45f, 53f, etc.

1 1. 4-1 5: The Johannine explanation of Lazarus' sickness and of why Jesus let him die — to make possible the miracle, to reveal the glory of God, and to confirm the disciples' faith (cf. 9. 3). 6 John's use of the passion prophecy will be discussed later. Here the thing to be noted is that his substitute for it, n.gf (as against Mk.

ΙΟ·33θ typically Johannine (= 9-40 and obviously intrusive (Bultmann, Johannes 304 ni ; Dodd, Historical Tradition 373^6) . The misunderstanding of a metaphor as an excuse for Jesus' explanation (verses 1 1 — 15) is a standard Johannine device for intro ducing secondary material (3.4; 4.11,33; 6.34,52; 8.22,33,39; etc.; Barrett, 1 73-1 74.

Bultmann's distinction of different types of misunderstanding — Johannes 304 n6 — is unimportant; this author was not so choosy.) Here Jesus' explanation is intended to prevent any discrediting of the miracle, which might result if κεκοίμητοα were taken literally and Lazarus supposed to have been merely cataleptic. Compare the addition of €l8ot€s otl άπεθανεν in Lk. 8.53 to prevent literal misunderstanding of the similar saying in 8.52 (= Mk. 5.39), and see below, on 11.17.

1 1. 16: The report of Thomas' devotion is an edifying addition (Barrett, 327 on verse 16; Bultmann, Johannes 305 n4).

1 1 . 1 7 : The specification that the body had been four days in the tomb is added to magnify the miracle and to refute any claim that Lazarus was merely asleep (see below, on 1 1.39b and 44; also Barrett, 335 on verse 39; Bultmann, Johannes 305 n6 — cf. ng). John likewise insisted that the blind man whom Jesus cured was born blind and that such a cure was therefore unheard of (9. if, 20,32, contrast Mk. 8.22f).John also made the centurion's slave into the son of a royal official (4.46, cf. Mt. 8.5; Lk.

7.2), made Jesus identify himself after the resurrection by showing his wounds (20.27), etc. These stories have grown with time.

1 1. 18: The precise specification of the distance from Jerusalem to Bethany is an other pseudohistorical explanation (it explains why the Jews came — they were so nearby; so Barrett, and Bultmann, Johannes , ad loc.). Thus it is probably late, rather than early, material, especially since it happens to be incorrect (Dalman, Orte 266).

1 1 . 19 : The "chorus " of Jewish mourners (so Dodd, Fourth Gospel 363) is introduced to provide additional witnesses to the miracle (Bultmann, Johannes 306) as well as

6. Against Barrett, 325 on verse 6, see Lightfoot, Jn. 219 ni, to say nothing of the text, ασθενεί, verse 3.

THE SECRET GOSPEL

the standard Johannine foil to Jesus. The mixed reactions of the Jews — skepticism, conversion, and talebearing (verses 37 and 45) — are also standard in Jn. (chs. 5-10 passim ).

1 1.20-27 : The homiletic conversation with Martha leading to the formal confession of faith is completely Johannine, though the confession is presumably that current in John's church (Bultmann, Johannes 308 n8; 309 n2).

11.28-31 : More "historical" explanation, to get the second sister and the chorus into the act (Bultmann, Johannes 309 n2; 31 1 n3). With Mary's arrival (verse 32)

John returned to the story as it lay before him, that is, to the saying /cupie, el ης a>8e

/c.r.A. from which he had departed (verse 21) to introduce the intervening sermonette.

(Cf. the similar case below, verse 34-37, and the note on 1 1.38. Mark does the same thing, 2.5b and 10b, evidently it was customary.) The parallelism with the longer text, which broke off at verse 21, now resumes. This indicates that the doubling of the sisters was John's work and was not in his source. He probably did it to make room for both Mary and Martha, whom he knew as a pair located in Bethany. The appeal KvpLe, el ης a>8e does not necessarily imply that messengers had been sent; it may be a typical expression of faith, later "explained" by the story of the sending

("•3)· 11.33: όν€βρίμησατο τώ πν€υματί καί erapa^ev ea υτόν. The difficulty of explaining this behavior as a consequence of the weeping of Mary and the Jews is indicated by Barrett's contortions (ad loc.). But καί όργ taOeis in the longer text is easily explicable as a consequence either of the woman's use of the messianic title or the disciples' rebuke of her. The disappearance of the appeal to the son of David (a title Jn. never uses) entailed the disappearance of the rebuke and left the anger unexplained. John

(or his source) therefore substituted the vague and portentous καί όν€βρίμησατο /c.r.A., which seemed suitable as an introduction to the miracle because of the words' magical overtones (Bonner, Technique 1 77ff; Lieberman, Tosefta Part V, p. 1363). Jn. regularly differs from the synoptics by its use of more pretentious language, with suggestions of something miraculous, mystical, or royal, e.g. : Mt. 8.5 || Jn. 4.46; Mk. 6.45ff ||

Jn. 6.i4f; Mk. 6.53 || Jn. 6.21; Mk. 8.29 |j Jn. 6.69; Mt. 18.3 || Jn. 3.3; Mk. 11.8 ||

Jn. 12.13; Mk. I4.43ff|| Jn. i8.3ff; Mk. 15.37 11 Jn. 19.29^ πώς eplXeL αυτόν is perhaps intended to show how the Jews twisted Jesus' innocent sorrow into evidence for a charge of homosexuality, ούκ eSvva το οΰτος 6 ανοίξας τους οφθαλμούς του τυφλοΰ πονησαι ΐνα καί οΰτος μη αποθάνη ' continues the theme by applying to Jesus a commonplace of Judeo-Christian polemic against pagan divinities and thus attempting to discredit his miracles; cf. Chrysostom, In Jo annem homiliae, ad loc.) Aristides, Apologia 11.3: Si igitur Aphrodite dea est et amatorem suum in morte eius adiuvare non poterat , qui alios adiuvare potest? Et ut audiatur naturam divinam in lacrimas <(cf. Jn. 11.36) . . . venire fieri non potest. (Again 11.5, on Rhea, and 6, on Kore.) The theme long continued popular; Wetstein on Jn. 11.37 quotes Ausonius on Zeus and Sarpedon. That it was applied to Jesus appears also from its use in the

crucifixion scene, Mk. 15*31 II Mt. 27*42, άλλους eacvaev, ea υτόν ου όυναται σώσαι, and in Midrash Tannaim on Dt. 3.23: "Before a man put his trust in flesh and blood

<i.e., in another man) and ask him to save him, let him <the proposed saviour) save

THE SECRET GOSPEL

himself from death first." The italicized words appear exactly in the Lucan parallel to Mk. 15.31 (Lk. 23.35). John's purpose in reporting the taunt here is the same as Mark's in the crucifixion scene — dramatic irony to emphasize (1) the coming resur rection, (2) the contrast between Jesus and the pagan divinities, and (3) the error of the Jews.

1 1.38: John again returns to his source by repeating the word at which he left it: ενεβριμησατο verse 33, πάλιν εμβριμώμενος verse 38 (cf. above, on 11.28—31). Again the parallelism to the longer text of Mk. stops with the first occurrence of the repeated word and resumes with the second. This all but demonstrates that the material be tween the two occurrences was added by John and was unknown to the author of the longer text of Mk. (Cf. Jn. 18.18 and 25, where the repeated words evidently came from a text like Mk. 14.54. The evidence of the longer text thus supports Bultmann's supposition of an interpolation between Jn. 18.18 and 25 — Johannes , ad loc. — against Dodd, Historical Tradition 82 ni.) 11.396-40: Another Johannine addition to magnify the miracle and explain its purpose; see above, on 11.4-15 and 17, and Bultmann, Johannes 31 1 nn4,6.

11.4^-42: This stage whisper to God — addressed as "Father" — (Barrett, ad loc.) is clearly an interruption in, the story and completely Johannine; cf. 5.36; 6.57; 7.29;

9.31 ; 12.28; 1 7. i ; etc; and Bultmann, Johannes 31 1 n6. Bultmann thought it obvious

("selbstverstandlich," ibid. 312) that the words of the prayer were not to be heard by the crowd, but Chrysostom was almost certainly right in treating them as public instruct 'on (De Christi precibus contra anomoeos IX end).

11.44: The grave clothes are another means of emphasizing that Lazarus had really been dead; see above, on 1 1.4-15, 17, and 38. It is not likely that John thought

Lazarus' moving, though bound, an additional miracle; the evangelist would not have been averse to throwing in a miracle, but did not visualize his scenes with suffi cient clarity to realize the difficulty. (Contra, Bultmann, Johannes 312.)

1 1.45-46: The Jews' reactions — unmistakably secondary in relation to the structure of the story, and typical of John, cf. 2.23; 7.3 1 fT ; i2.iof, 42, etc.; Barrett, 337. i i.47-54a: An independent tradition, developed by independent invention. Its use in relation to the larger structure of the Gospel is obvious (Barrett, 337) ; it is inserted at this point to provide John's regular explanation for a withdrawal by Jesus, since a withdrawal was reported by his source (11.54b || the longer text) and had to be

"explained." Cf. Jn. 7.1; 8.59; 10.39; etc· This completes the list of material in Jn. which is unparalleled in the longer text of Mk. Every bit of it is obviously secondary and obviously Johannine. With the above analysis that of Dodd, Historical Tradition 228ff, can now be compared. Without knowing the longer text, Dodd concluded that the Johannine account was a reworking of an earlier story of synoptic type, which, however, he thought it impossible to dissect. He remarked — 228 n2 — "as Johannine traits. . . the identification of in dividual characters, the measurement of time and space (two days, four days, fifteen

stades), the use of the term ol Ιουδαίοι. Locutions with a Johannine ring are ύπ ερ της δόξης του θεοΰ, ΐνα δοξασθη 6 νιος του θεοΰ , παρρησία, χαίρω δά υμάς ίνα πιστευσητε, οσα αν αίτηση τον θεόν δώσει σοι (cf. 16.23), €Ψι> €LS τ°ν α^ώνα (twelves times in

THE SECRET GOSPEL

John, twice in Mark, once each in Matthew and Luke), ό εις τον κόσμον ερχόμενος, οφη την δόξαν τον θεόν, ΐνα ττιστενσωσιν ότι σν με άττεστειλας ." The agreement of this list with the results of the above anslysis is clear. Accordingly there can be no question that the story in the longer text of Mk. is more primitive in form than the story of Lazarus in Jn. Further, it is impossible to suppose that the author of the longer text of Mk. used, or even knew, the Johannine Lazarus story. Had he known it, his text would certainly have shown at least some of the secondary Johannine traits listed above. Since it has none of them, it must be completely independent of JnThese facts make it possible to distinguish, in the Johannine Lazarus story, the source John used. It is often recognizable by its parallels to the longer text. But in the material paralleled there are important differences: (i) The sisters are represented as sending word to Jesus (above, on 11.3 and 32). (2) The sister appeals to Jesus with the cry κνριε εΐ ης ώδε κ.τ.λ., rather than νίε Δαβίδ, ελεησόν με (on 11.28—31).

(3) οργίζομαι is replaced by εμβριμάομαι (on 1 1.33)· (4) No φωνή μεγάλη is heard from the tomb on Jesus' approach. (5) The stone is removed by persons unspecified, not by Jesus himself. (6) Jesus calls Lazarus forth from the tomb ( φωνή μεγάλη) instead of going in and raising him by hand. (7) The Johannine story concludes with reference to Lazarus' grave wrappings and Jesus' order to untie him (on 11.44), whereas the story used by the longer text said nothing of these, but concluded with the raising.

( εμβλεφας αντω ηγάπησεν αυτόν κ.τ.λ. probably comes from the editor, and καί εξελθόντες κ.τ.λ. is an editorial transition to the next episode.)

Not all of these differencescan confidently be attributed to John's source. As indicated above, the sisters' sending to Jesus probably was John's work; likewise the replacement of οργίζομαι by εμβριμάομ at. The same probability can be estab lished for the change in the use of φωνή μεγάλη (4 and 6, above). The cry from the tomb would have led many ancient readers to question the miracle. Had the man who was raised really been dead ? Stories of persons who were thought to have died but came back to life were frequent in antiquity (Plato, Republic 614b; Proclus, In Platonis rem publicam , ed. Kroll, II.113; Kerenyi, passim', Philostratus, Vita Apollonii

IV. 45, Eus., Against the Life of Apollonius 26 and 31), and the notion that the persons

"raised" by Jesus had not been really dead was a frequent embarrassment to Christian apologists (Origen, Contra Celsum II. 48; GCS, Origenes, vol. 12.III.1, frags. 185-186;

Ephraem, Commentaire VII. 27 p. 77; Cramer, I.321 on Mk. 5.43; Chrysostom, In Matthaeum homiliae 31 on Mt. 9.i8ff). We have seen above that this embarrassment was already felt by John and that (as Chrysostom remarked) John emphasized the four days' entombment and the smell, and so on, "that they should not have any ground to disbelieve that the man whom (Jesus) raised had been dead." Given this apologetic concern, it is understandable that either John or his source should have suppressed the voice from the tomb and transferred the φωνή μεγάλη to Jesus — in spite of the fact that it is somewhat out of character, as Cyril of Alexandria remarked

(In Johannem, ad loc.).

The original significance of the cry from the tomb is probably indicated by the use of φωνή μεγάλη in Mk., where it occurs often at crises in the relations between

THE SECRET GOSPEL

spirits and men— see the commentary on III. i , above. In 1.26 the demon φωνησαν φωνή μεγάΧη έξήΧθεν; in 5.7 the legion of demons κράξας φωνή μεγάΧη respond with a counter spell to Jesus' command that they leave their victim; in Mk. 15.34 Jesus himself εβόησεν . . . φωνή μεγάΧη ΕΧωι EX ωι Χαμα σαβαχθανει, and in 15.37 <*</>€1? φωνήν μεγάΧην εξεττνευσεν (cf. Fenton, Destruction 57)· In the longer text, accordingly, the φωνή μεγάΧη is probably the cry of Death, departing from its prey: cf. Mk. 1.42;

Lk. 4.39; I Cor. 15.26; Apoc. 6.8; 20.13^ Brandon, Personification 330b Death or Hades releasing the soul of Lazarus appears in figured representations of the miracle from the fifth century on, Millet, Recherches 233. Reau, Iconographie II.ii.338, calls this

"the first version of the miracle properly so-called." For these two references I am indebted to Meyer Schapiro. [Substantially this same interpretation was proposed independently by R.S., who also, with T.B., suggests that there may be a connection between the longer text of Mk. and Jn. 12.17, where the crowd accompanying Jesus into Jerusalem celebrates his miracles by declaring τον Λάζαρον εφώνησεν εκ του μνημείου καί ηγει ρεν αυτόν εκ νεκρών. This could have been derived from the longer text by sub stitution of τον Λάζαρον for 6 Λάζ αρος either deliberately or by misreading.

The possibility of misreading, especially of an Aramaic text, can be seen from the Peshitta, where Jn. 12.17 reads k<Ljo Lea y» y» Were it not for the doubling of the first y this might represent 6 Λάζαρος εφώνησεν εκ τοΰ μνημείου, καί

<(ό Ίησοϋς'} ηγει ρεν αυτόν εκ νεκρών, where the sequence of events and the underlined words are paralleled exactly — albeit with interruption— in the longer text.] Such a misreading (substitution of τον Λάζαρον for ό Λάζαρος) would be psychologically likely, since Lazarus was dead and therefore not expected to call out. And the demonological parallels given above argue that the story of the raising in the longer text of Mk. is primitive. The fact that it contains a difficulty, which the story in Jn. does not, also argues that it is older than the Johannine form: difficilior lectio. Accord ingly, I think John knew the story in a form similar to that of the longer text, and the transference of the φωνή to Jesus should be attributed to him rather than to his source.

We saw above that it accords with his apologetic concerns. (The possibility that John's source was written in Aramaic will be of some importance hereinafter.)

Thus a number of the peculiarities of the Johannine story are to be referred to John, along with the obviously Johannine interpolations previously listed. However, there is at least one difference which can more confidently be referred to John's source — the appeal κύριε, ει ης ώδε, κ.τ.Χ., which John repeated as a catchword in verse 32 when he came back to his source after his homiletic and explanatory excursion in verses 23-31 (see above, on 1 1.28-31 and 38). Similarly, the differences as to who moved the stone (no. 6, above) and how the youth was raised (no. 7) are probably due to John's source; at least, neither of them is directly accounted for by John's peculiar interests or style.

These differences which can be referred to John's source afford reason to believe that even the source was later in form than the story in the longer text.

In the first place, the appeal κύριε, ει ης ώδε, ουκ αν άττεθανεν 6 άδεΧφός μου, is a cry of grief from the Church after Jesus' departure and an expression of hope in a future resurrection, whereas υίε Δαβίδ, εΧεησόν με expresses the primitive Palestinian

THE SECRET GOSPEL

hope for immediate action by a present, Davidic Messiah (and was therefore dropped when Christianity moved away from its Palestinian Jewish origins).

In the second place, that Jesus himself should move the stone is more likely to be primitive than that others should do it for him. So far as I can recall, Jesus is never reported in the Gospels to have done any hard manual labor — except for Jn.'s emphatic "himself carrying his cross" (19.17), which is the last humiliation before crucifixion. This lack of reference to Jesus' doing any manual labor is presumably an attempt to make him respectable, like Matthew's alteration and Luke's omission of Mk.'s 6 τεκτων (Mk. 6.3 || Mt. 13.55; Lk. 4.22). The quite casual and nondogmatic way (contrast Jn. 19.17) with which the longer text refers to Jesus' removing the stone makes it seem that we have here early material.

Finally, it might be argued on similar grounds that a story which reports the direct, physical method of taking the hand and literally raising the dead is probably more primitive than one which reports a raising by remote command. It was re marked above (commentary on II 1. 4) that references to χειρ as an instrument of supernatural help are substantially more frequent in Mk. than in the later synoptics and Jn. (10 or 11 in Mk., 7 in Mt., 5 in Lk., o in Jn.) Notice also the disappearance from Mt. and Lk. of Markan miracles worked by physical means (7.32-37; 8.22-26) and the continuation of this tendency in later Christian apologetics (Fridrichsen,

Probleme 61). Is it possible that the Johannine story was also influenced by considera tions of purity, to eliminate the reference to Jesus' touching a corpse?

Having thus established grounds for belief that John's source was later than the resurrection story in the longer text of Mk. and differed from it substantially, we can conclude that Jn. was independent of the longer text no less than it of Jn.

G. The order of events in Mk. and Jn.

This conclusion gives particular importance to the parallelism in order of events which appears between the latter halves of Mk. and Jn. once the longer text is put in its place in Mk. This parallelism can best be demonstrated by an abbreviated synopsis :

Jn. 6.1 μετά ταύτα άπηλθεν 6 Τησοΰς πέραν Mk. 6.32 καί άπηλθον εν τω πλοίω είς έρημον τόπον κατ ίδιαν. της θαλάσσης της Γαλίλαίας της Τίβεριάδος.

— The feeding of the five thousand The feeding of the five thousand 6.45 καί ευθύς ηνάγκασεν τούς μαθητάς αυτού

6.16— 17a ώς δε όφ ία εγενετο, κατεβησαν οΐ μαθηταϊ αυτού επί την θάλασσαν , καί εμβηναι είς το πλοΐον καί προάγείν είς το πέραν προς Βηθσαϊδάν , εως αυτός απολύει εμβάντες είς πλοΐον ηρχοντο πέραν της θαλάσσης είς Καφαρναουμ. τον οχλον.

THE SECRET GOSPEL

— 6.46 καί άποταζάμενος αύτοις άπηλθεν εις το (6. 1 5 Ιησούς ούν γνούς οτι μέλλουσιν έρχεσθαι καί άρπάζειν αύτον ίνα ποιησωσιν βασιλέα , όρος προσεύξασθαι. ανεχώρησεν πάλιν εις το ορος αύτός μόνος.)

= The walking on the sea The walking on the sea 6·54-55^ καί έζελ θόντων αυτών εκ του πλοίου ευθύς έπιγνόντες αυτόν περιέδραμον όλην 6.24—^ 25a οτε ούν είδεν ό οχλος ότι ' Ιησούς ούκ εστιν εκεί ούδέ οι μαθηταί αύτοΰ , Omitted by Luke ένέβησαν αυτοί εις τα πλοιάρια καί ηλθον την χώραν. εις Καφαρναουμ ζητοΰντες τον Ίησοΰν. καί εύρόντες αύτόν . . .

= Discussion: Jesus is the bread of life Summary : Jesus' miracles of healing The dispute on handwashing Trip to the territory of Tyre The Syrophoenician Return to Galilee The dumb man ( εφφαθα ) The feeding of the four thousand The demand for a sign The saying on the leaven of the Pharisees

— The blind man of Bethsaida = 6.66-69 Peter's confession (in Caper naum ?) 8.27-30 Peter's confession (in Caesarea Philippi) Peter is Satan = Judas is a devil The sayings on self-sacrifice The transfiguration The demoniac boy 9.30-31 κάκεΐθεν έξελ θόντες παρεπορεύοντο διά της Γαλιλαίας, καί ούκ ηθελεν ΐνα τις

7.1 καί μετά ταΰτα περιεπάτει ό ' Ιησούς εν τη Γαλιλαία · ού γάρ ηθελεν εν τη ^ Ιούδα ία γνοΐ· έδιδα σκεν γάρ τούς μα θήτας αύτοΰ, περιπατεΐν, ότι έζητουν αύτόν οι ' Ιουδαίοι άποκτειναι. καί έλεγεν αύτοΐς ότι 6 υιός τού ανθρώπου παραδίδοται εις χεΐρας ανθρώπων καί άποκτενοΰσιν αύτόν . . .

Jesus' brothers taunt him The dispute on precedence The stranger who exorcized The sayings on scandals 7.IO ώς δε άνέβησαν οι άδελφοί αύτοΰ εις την

IO. I a καί έκεΐθεν άναστας ερχεται εις τα όρια εορτήν, τότε καί αύτός άνέβη . . .

The disputes in Jerusalem The man born blind της Ίουδαίας The sayings on the door to the sheep The appeal to the witness of his works IO.lb καί πέραν τού Ίορδάνου, καί συνπορεύIO.40— 41a καί άπηλθεν πάλιν πέραν τού

Ίορδάνου εις τον τόπον όπου ην ' Ιωάννης τό πρώτον βαπτίζων, καί 'έμενεν εκεί, καί πολλοί ηλθον προς αύτόν.

THE SECRET GOSPEL

The question on divorce The blessing on children The rich young ruler The sayings on scandals The preface to the Lazarus story 1 1.7—8 επειτα μετά τούτο λεγει τοΐς μαθηταΐς,

IO.32 ησαν δε εν rfj όδώ άναβαίνοντες είς άγωμεν είς την Ίουδ αίαν πάλιν, λεγουσιν

'Ιεροσόλυμα, καί ην προάγων α ντους ό αυτώ οι μαθηταί, Ραββε ι, νυν εζητουν σε ' Ιησούς, καί εθαμβοΰντο, οί δε άκολουθοϋντες λιθάσαι οί Ιουδαίοι , καί πάλιν υπάγεις εκεί ; εφοβοΰντο.

Jesus' announcement of Lazarus' death Jesus' prophecy of his own passion and resurrection and prophecy of his resurrection The Lazarus story Longer text: The Lazarus story The nocturnal initiation The Jews' reaction and plot (The Jews' plot, Mk. 14. 1-2, infra) III. IO— I I εκε ΐθεν δε άναστάς επεστρε φεν είς

1 1 .54 6 ούν Ίησοϋς ούκετι παρρησία περιεπάτει εν τοΐς Ίουδαίοις, άλλα άπηλθεν το πέραν του Ίορδάνου. εκεΐθεν είς την χώραν εγγύς της έρημου , είς

Mk. The question of James and John !7) Longer text: The events in Jericho Mk. Bartimaeus infra) The entry of Jerusalem The cursing of the fig tree (The entry of Jerusalem, Jn. 12. 12-19, The cleansing of the Temple The fig tree found withered (The cleansing of the Temple, Jn. 2.132.18) (The question as to Jesus' authority, Jn.

The question as to Jesus' authority The parable of the rented vineyards Questions by Herodians, Sadducees and a scribe The question as to the son of David The widow's mite The prophecy of the destruction of the (The prophecy of the destruction of the Temple Temple, Jn. 2.19-22) The prophecy of the end (The Jews' reaction and plot, Jn. 11.47The Jews' plot

54, supra) The anointing in Bethany The anointing in Bethany The entry of Jerusalem The request of the Greeks (The entry of Jerusalem, Mk. 11.1-10, supra) The Evanglist's comments on the Jews Jesus' declaration of his mission The preparation for the last supper The footwashing The last supper The last supper The passion story The passion story

THE SECRET GOSPEL

In this synopsis the things which require explanation are the continued parallelism of the geographical framework (shown by the verses quoted in Greek) and the near identity in order of those larger elements which the two Gospels have in common.

Of these elements (indicated by the equal sign) three— the cleansing of the Temple, the question as to Jesus' authority, and the prophecy of the Temple's destruction — are not properly in question, since they appear in Jn. 2. (However, it is interesting to note: [1] that in Jn. 2 they appear in the same order as they do in Mk. 1 1 — 1 3 ; [2] that the Streitgesprache, with which they are closely connected, are in Mk. divided between 2. 1-3.6 and 11. 27-12. 37. These facts suggest that their material derives in part from an independent block of tradition which had some connection with an early stage in Jesus' career — a suggestion to which we shall return later.) Of the other 12 major elements listed as common to Mk. (including the longer text) and Jn., all occur in the same order, save that Jn. has the entry of Jerusalem after the Jews' plot and the anointing, and Mk. has it before them.

This coincidence in order of so many events can hardly be accidental. Yet, it seems unlikely that John used Mk. or Mark, Jn. (Haenchen, Probleme; numerous studies of detail by Buse; more recently Smith, Jn. 12.12; and above all Dodd, Historical Tradi tion). And we have already seen evidence that the longer text and Jn. were independ ent developments of a common source, possibly in Aramaic, from which Jn. was separated by at least two removes (since his immediate source was later in form than the story in the longer text). Evidence for the independence of Jn. and canonical Mk. is to be found in the great differences of form between some of the elements common to them: Peter's confession, the curse on Peter ("Simon") or Judas ("the son of Simon"), the anointing, the last supper, and the passion story. Most of these were mentioned above as "remote, but unmistakable parallels, which can best be explained as divergent forms of the same tradition."

The same relation seems to hold between the elements of the geographical frame.

They are undeniably parallel, and in outline their accounts of Jesus' movements are substantially identical.7 But the verbal coincidences between them are trivial ( άπηλθον , τούς μαθητας αυτού, εμβηναι εις π λοΐον, πέραν, εις το ορος, της Γαλιλαίας, ονκ ηθελεν, άποκτενοϋσιν, etc.) and the differences, not only of wording, but also of content, are so substantial that it would be implausible to suppose either author got them from the other (especially since material of this sort is for the most part theologically un important and therefore not likely to suffer deliberate changes). Therefore their continual parallelism and substantial differences must be explained by supposition of a common source of which both authors used different developments.

7. Mk. IO. I, καί έκε ίθεν άναστάς έρχεται εις τα όρια της Ίουδαίας καί πέραν τον ' Ιορδάνου , has often been thought corrupt, or "explained" as meaning the reverse of what it says. However, neither treatment is necessary, and the plain sense of the verse as it stands (he went first to Judea and then to Transjordan) is supported not only by the parallels in Jn., but also by the independent tradition in Lk. 9.5 iff and 17.11. See the discussion in Taylor, ad toe. [C.R. suggests that the text of Mk. 10.1 as given above may have been produced by abbreviation, that is, by omission of the stories ol what hap pened in Judea. He compares Mk. 7.31, where the strange geography (from Tyre through Sidon to the Sea of Galilee in the midst of the Decapolis) is possibly the result of amalgamation of parts of a number of introductory notices.] l6l

THE SECRET GOSPEL

The same conclusion is indicated again by the fact that in both Gospels the parallel episodes stand in the same relation to the parallel framework; that is, the same events occur not only in the same order, but also in the same places in the parallel frames, and are thus for the most part located in the same geographic places. Here again there are discrepancies sufficient to make it unlikely that either author used the other, but insufficient to obscure the basic identity of the outlines.

Finally, the similarities demonstrated by the above synopsis would be increased yet further if we were to accept the theory of Dodd [Close) and Huffmann (

Sources 128) that Mk. 8.1-26 and Mk. 6.30-7.37 are variant forms of the same body of tradition. This theory yields for Jn. 6.1-65 a double set of Markan parallels which cannot plausibly be explained as accidental coincidence of editorial constructions, especially because the coincidence lies less in matters of wording (which might be editorial formulas) than in the order of events which are not obviously identical but turn out, one after another, to be basically similar. Discussion of Dodd's theory would be irrelevant for our present purpose, but his evidence strengthens materially the already strong case to be made from the synopsis printed above. Incidentally, although when he wrote Close Dodd was of the opinion that his data were evidence of John's use of Mk. (p. 288), he has since changed his mind and treated them as evidence of the dependence of both Gospels on common tradition (

Fourth Gospel 448ff; Herrnworte and Historical Tradition , passim ).

It seem.? unlikely, however, that oral tradition should account for so extended an agreement in order — particularly since the events and, even more, the geographical references reported do not seem to be connected by any coherent plot which would fix their order in the narrator's memory. If we therefore suppose a common written source, how can we account for the differences between the Gospels, and especially for the differences between the material they have in common (which probably came from the source) ? The source may have been in Aramaic and the differences may result in part from different translations. Into these different translations both Mark and John would then have inserted, chiefly from other sources, the additional material peculiar to their own Gospels. These suggestions obviously resemble those of Dodd,

Framework , on which there have been many attacks — for example, Nineham, Order,

Robinson, Quest 486!, and Trocme, Formation 23ff. See Dodd's reply to Nineham in Historical Tradition 233 n2. In the same book, 235ff, Dodd picked out of Jn. a number of transitional passages which he thought derived from material akin to that of the synoptics. Of these, three (Jn. 7.1-2; 10.40-42; 11.54) occur after 6.1, where the parallelism which we have observed begins. All three of these can now be seen to be paralleled in Mk. (1 1.54 in the longer text). Clearly, this does not settle the matter.

But the new evidence is strong prima facie support for Dodd. And the question is important. If a good part of the content of a common source can be discovered by comparison of Mk. and Jn., we shall have not only some notion of an extremely early Gospel, but also good indications of the peculiar elements of the Markan and the Johannine traditions, as evidenced by their additions to this source.

That some of the Markan insertions were made late in the development of Mk. is suggested by the coincidence ofjn. with Lk. in most of Lk.'s "great omission" of

THE SECRET GOSPEL

Mk. 6.46-8.26. We need not, however, suppose that either John or Mark copied his source whole. Each may have omitted or reworked parts of it. In Jn. 11.7-16, for instance, it is plausible to suppose that John reworked the incoherent elements of the source (reflected by canonical Mk. and the longer text) into a coherent preface to the resurrection of Lazarus. Reworking of the Johannine preface into the incoherent material now found in Mk. and the longer text is incredible. Again, neither Jn.'s omission of the nocturnal initiation nor the longer text's inclusion of it is proof that it was not — or was — in their common source.

But the facts that both Jn. and the longer text of Mk. do include the resurrection story, that both locate it at the same place in their outline, and that both introduce it and follow it by similar pieces of framework — these facts make it likely that the resurrection story was part of the common source on which both Mk. and Jn. were dependent. But if the story was in Mk.'s source it was probably in the earliest form of Mk. If so, we should suppose that the canonical text of Mk. was produced, at least in this instance, by abbreviation of the earlier, longer text. We shall have later to weigh this conclusion against the opposite one, reached above from consideration of the stylistic evidence.

Before leaving the relation of the Johannine Lazarus story to the resurrection story in the longer text, it should be noted that both have a number of parallels to the stories of Jesus' resurrection. This is easily explicable. The similarity of content would necessitate some, chance might account for others, the tendencies to assimilate the phrasing of similar stories and to regard Lazarus' resurrection as a prefiguring of Jesus' would produce yet more. Accordingly it is not surprising that a tomb located in a garden and closed by a stone which had to be rolled away should appear in the stories of both resurrections. (As we saw above on Jn. 1 1.37, John also included in his Lazarus story the same sort of polemic material which Mk. put in his crucifixion scene, and for the same purpose — dramatic irony before the resurrection.)8

8. But there is one further parallel between Jn. and canonical Mk. which deserves attention: Jn. insists that Lazarus was raised on his fourth day in the tomb (11.17,39, emphatic repetition). The passion prophecies of Mk. 8.31, 9.31, and 10.34 date Jesus' resurrection μετά τρεις ημέρας, which might mean "on the fourth day." This is why the date was changed to τη τρίτη ημέρα by Matthew in all instances and by Luke in two. (In the third — 9.44— Luke eliminated the date entirely. Note also the correction by the copyists of Mk., Williams, Alterations 45.) For the meaning of the expression in Mk. cf. Mk. 9.2, where μετά ημέρας εξ presumably means "on the seventh day" — as shown by the parallel in Ex. 24.16 — no doubt for Sabbatarian or numerological reasons (Lohmeyer, Grundmann). LXX uses μετά with a number of days to indicate the day following the number given. [C.R. remarks that this is particularly clear from the parallelism in Hosea 6.2: ύγιάσει ημάς μετά 8υο ημέρας, έν τη ημέρα τη τρίτη άναστησόμεθα.] (On this the third-day resurrection tradition may have been built.) See also Dan.

1. 15 (LXX), with which cf. Theodotion; Gen 7.10; 8.3,6, etc. Luke seems to have understood Mk.'s μετά ημέρας εξ in 9.2 as meaning "on the seventh day," since his ωσεί ημέραι οκτώ (9.28) are probably the 7 days of Mk. plus 1 for the day from which the count began (so, on Mk., Swete, followed by Lagrange and Klostermann; on Lk., Creed, followed by Luce; contra, Rengstorf). Later on, however, both Mk. (chs. 15-16) and Jn. (19-20) date the resurrection on the third day after the crucifixion.

Therefore the expectation of a resurrection on the fourth day is so odd that its appearance in the Markan passion prophecies and the Johannine Lazarus story may be thought evidence for the hypo thesis that John used a source like that of Mk.-plus-the-longer-text, in which the Lazarus story was closely connected to a resurrection prophecy of the Markan type. The only (?) other traces of the

THE SECRET GOSPEL

D. Relation of the new material to the structure of Mk.

Now we turn to the evidence of Mk. itself as to whether or not the canonical text is an abbreviation of the longer one.

I. POSITION IN THE "HISTORICAL OUTLINE"

The longer text reports that the dead man's sister addressed Jesus publicly by the messianic title "son of David" and that the disciples rebuked her. This is an example of the Markan motif of the "messianic secret" which is always being let slip and then hushed up— the closest parallels are in i.24f; 3.1 if; 8.2gf; 9.9,306; 10.48; a recent discussion is Burkill's Revelation esp. 62ff. In canonical Mk. the title "son of David" first appears in 10.47 (the Bartimaeus story) and is difficult to explain there. People had not been saying that Jesus was the Messiah (Mk. 6.i4f; 8.28). Only Peter had guessed it (8.29) ; and he and the others who heard him had been warned to keep it secret (8.30). How, then, did the title indicative of this secret get into the mouth of the beggar Bartimaeus, outside Jericho (10.47) ? This question was asked by Ebeling

Messiasgeheimnis 92) with the confidence that it would be historically insoluble. The longer text does not supply a historical solution, but it does present a sequence of facts from which historical imagination can create an understandable sequence of events. For between Peter's confession and Bartimaeus' appeal it puts first the use of the title by one of the women of a family with which Jesus was intimate (Jn. 11.5) and then a visit by this woman to Jericho, where (or before which) she had some sort of difference with Jesus such that he did not "receive" or "welcome" her and her companions. By the time he left the city, even the beggar by the roadside knew he claimed to be the son of David.

This argument might have been well received fifty years ago. Today the reader will object: (1) Whatever may be thought of the primitive outline of Mk., the present order of events is not in detail historically reliable (Schmidt, Rahmen). (2) The intro ductory inventions in Jn. 1 1.1-5 are unreliable (see the comments above) and cannot be used to prove Jesus' intimacy with the family of the deceased; in the longer text of Mk. the sister appears without any introduction and her usage of the title "son of David" is no more explicable than Bartimaeus'. (3) A likely "historical" explan ation which would account for a given sequence of events is not necessarily true. To demonstrate its truth one would have to demonstrate that no other explanation was equally likely and that what did happen was what was most likely to happen (often not the case). Such demonstration is impossible in the present instance, therefore this possibly historical construction is of little value as evidence for the question of whether the longer text was prior to the shorter one or vice versa. expectation are Mt. 1 2.40; 27.63 ; and the western text of Acts 10.40 — another example of the corruption of the western text by the influence of this section of Mk. (Stendahl, School , has argued that Mt. 12.40

is a later insertion. Trocme, Formation 180 and n24, also remarked the difficulty and drew from it the conclusion that the original Mk. did not contain chs. 14-16. But if a later interpolator could overlook such contradictions, so could an original compiler. The material is better evidence for diversity of tradition than for details of literary history.)

THE SECRET GOSPEL

2. PARALLELS TO THE TRANSFIGURATION AND PASSION STORIES

More important is the fact that insertion of the quotations from the longer text into canonical Mk., at the places Clement indicates, enables us to present three important sections of Mk. as parallel constructions showing the same basic pattern.

Mk. 10.20: A man's pro Mk. 8.29: Peter's confession of faith fession of good works.

10.21 : Jesus loved the man (Mt. 16.17: Jesus' blessing (Lk. 22.31 : Jesus' prayer for of Peter.) (a youth, Mt. 19.20).

Peter.)9 Mk. 14.27: Jesus' prophecy 10.21: Jesus' demand, For sake all and follow me.

Mk. 8.31: Jesus' prophecy of his passion. of his passion.

10.22: The man rejects 8.32: Peter rejects Jesus' teaching (an expression of 14.29: Peter rejects Jesus' teaching (an expression of his love for Jesus).

Jesus' teaching. his love for Jesus).

8.33 : Jesus calls Peter Satan.

10.23: Jesus declares that the man can hardly enter 14.30: Jesus' prophecy that Peter will deny him. the kingdom.

10.28: Peter's claim, We 14.31: Peter's promise, Though it cost me my life, I shall not deny you.

8.34: Jesus' demand, Deny yourself and follow me , Lose have forsaken all and followed your life for me.

10.32: Jesus' prophecy of 8.34: Implies Jesus' passion.

1 4. 53 if: Jesus' passion. his passion.

Longer Text: Resurrection you.

(Supposed but not reported:

9.1: Jesus' prophecy, Some shall not taste death.10 of the man (a youth). His Jesus' resurrection.) love for Jesus.

LT : Teaching of the mystery 9-2ff: Transfiguration ap pearance (in white) to 16.5: A youth in white ap σαββάτον.11 pears, anouncing the res to the youth (in white).12

Peter, James, and John. urrection.13

9. Note the similarity between Lk.'s prayer for Peter (that Satan shall not prevail over him, that he shall strengthen the brethren) and Mt.'s blessing of Peter (the Church shall be built on him and Hell shall not prevail over it). Furthermore, Lk.'s prayer follows Jesus' teaching that greatness is service and the promise that in Jesus' kindom the twelve shall sit on thrones. This is akin to the request of the sons of Zebedee to sit at Jesus' right and left hands in his kingdom and Jesus' reply that greatness is service.

And this request and reply immediately follow the second passage shown here (Mk. 10.20-34 plus the longer text).

10. This verse is probably to be connected with the Lazarus story and the expectation that the beloved disciple would never die, an expectation also attributed to events in a resurrection appearance

(Jn. 21.23). Eckhardt's denial of the relationship ( Tod i6f) does not allow for the fluidity in form of traditional material nor the importance of the common tradition (demonstrated above) behind Mk. and Jn.

11. In its present Markan context this cannot mean μεθ' ημέρας εξ; yet σάββατον is used in the very next verse to mean "week," and both Matthew (28.1) and Luke (24.1) so changed the construction as to eliminate the ambiguity. Is it accidental that the rejected sense should accord so closely with the two parallels?

12. [K.S. suggests that Mk. 10.35-45 is related to 10.32-34 as is 8.32-33 to 8.31.] This I doubt.

There is no rejection of Jesus and no curse on the apostles.

13. Note that Mk. 16.7 promises a resurrection appearance, presumably with some teaching (secret?).

It may or may not be supposed that this was the appearance to Peter (Lk. 24.34; I Cor. 15.5) of which the disappearance is one of the most suggestive mysteries of early Christian tradition. (The attempted solution by Annand, He Was Seen, does not convince me.)

THE SECRET GOSPEL

These texts seem to follow the same basic pattern : acceptance, blessing, demand for sacrifice, rejection, curse; renewed demand for sacrifice, Jesus' fulfilment of the demand, resurrection, and, one week later, revelation of the mystery.14 Yet it would seem also that the expression of this basic pattern has been carried out with consider able differences even as to the most important points: Mk. 8.34 does not explicitly foretell or report Jesus' passion, but merely implies it, and Mk. 16. iff merely implies the resurrection; but the implication is certain in both instances, so there is no denying the identity of the basic pattern. Similarly, in 8.29-9.10 and in 10.20-34 plus the longer text, the pattern is presented in concentrated form with relatively little extraneous material, whereas in 14.27-16.8 it is worked into the much larger passion story in which the earlier parts of it (14.27-31) appear as an excrescence unimportant to the course of the main action.

Thus the relation between these passages is somewhat similar to that between the variant forms of identical stories (discussed above) and the variant developments by Mark and John of a common narrative presumably given by tradition. But here there seems to be an important difference. It can hardly be supposed that the trans figuration, the Lazarus story, and the passion story are all variants of a single narrative. (Against the supposition that the transfiguration is an account of a resurrec tion appearance see above, p. 149). Nor do the stories of Peter's confession and the rich young ruler seem variants of a single original. Rather, the common pattern here is the result not of common origin, but of common purpose, which has three times put together pieces of different origins to express the same teaching. The teaching seems to be that of human depravity and the consequent vicarious atone ment: Even those who accept Jesus (or the Law) and therefore receive his blessing cannot accept the demand for sacrifice, and therefore come under his curse; conse quently he sacrifices himself, rises from the dead, and communicates to his unworthy followers the mystery of his resurrection. The parallels to this in Paul are well known

(e.g., Rom. 5.8: God shows his love for us in that while we were yet sinners Christ died for us), and the Hodayot from Qumran have shown that even before Jesus' time some circles in Judaism were magnifying human depravity and God's salvation of his elect in spite of it ( Hodayot I.2off; III.24ff; IV.29ff; etc.).

However, in none of the three passages outlined above is the doctrine clearly ex pressed. Instead it is implied by the sequence of stories and sayings, of which no single one states it clearly. Only when the three pericopae are put side by side does the repetition of the pattern and thus its significance appear. The pattern can hardly be accidental — the parallels are too close; why, then, did the author not state its significance? Evidently he expected his book to be taught, and left the explanation of these patterns to the teacher. This is exactly what Clement's letter says Mark did

(I.22-26). The same conclusion was reached by Bird (γάρ 174).

But the fact that Clement realized that there was latent teaching in Mk. does not prove that his account of the origin of the longer text is correct. And the above

14. K.S. comments: "Another complex is Thomas in Jn. 20 with a week's delay, coupled with Thomas' role in the Gospel of Thomas, where he receives the secret revelation in three words." This is perhaps important, especially because of the connection of Thomas with Salome, to be noted below.

THE SECRET GOSPEL

demonstrated fact — that the longer text taken together with Mk. 10.20-32 embodies a pattern found in two pericopae of the canonical Gospel — suggests that the longer text was originally part of the Gospel, since the patterns are presumably the work of a redactor and the three instances of the same pattern look like three examples of the work of the same redactor. This is particularly so because the patterns are not explicit or even obvious; therefore one might say they would not have served as models for some later editor who added the material of the longer text to produce another example. But against this there is evidence that these patterns were recog nized by later editors (who had the exegetic traditions of their churches to guide them). Matthew added to Mk. the blessing of Peter, Luke, the prayer for Peter, and both of these additions stand in exactly the right places to extend and emphasize the patterns; this can hardly be accident. (See, further, note 9 on the three texts out lined above.) Accordingly, it is not impossible that the material of the longer text should have been interpolated after 10.34 in order to produce another example of a pattern useful for exegetic purposes. However, the supposition that all three examples were produced by the same redactor seems more likely. The additions in Mt. and Lk. are comparatively minor and — although they embody independent tradition — their attachment to the pattern would have been suggested by its existence. In Mk. 10.2032, on the contrary, the most important elements of the pattern are lacking; and the addition of the material of the longer text would not merely have filled out an existing example, but would have created a new one.

Postscript: On rereading this section in 1966, some three years and more since it was written, I must confess I am more dubious than ever, not only as to the significance of these parallels, but even as to their actual parallelism. Mk. 10.21 is not actually paralleled in Mk. 8 or 14, and even if the additions from Mt. and Lk. be supposed to have been derived from early tradition, love is neither a blessing nor a prayer.

The rejection of Jesus in Mk. 10.22 is something quite different from Peter's expres sions of loyalty in 8.32 and 14.29, and they, in turn, are different from each other.

10.28 is not a new demand for sacrifice, but a claim to have made the necessary sacrifice. In the longer text and in 9.1 we have the youth's resurrection, not Jesus'.

Consequently — though the one resurrection was doubtless thought to prefigure the other — the proposed interpretation in terms of the vicarious atonement seems forced.

A further fact which seems to me important is that these alleged parallels do not fit easily into the structure of Mk. revealed by the interpretation of Mk. 10, to be pre sented in the following section. But the interpretation of Mk. 10 is much better supported than the argument for the above parallels. Accordingly, I suspect the parallelism may be due not to "Mark's" intention but to my invention. But if I was deceived before, when I could see it, I may be deceived now, when I can't. According ly, I submit the problem to the reader.

3. RELATION TO THE BAPTISMAL CONCERN OF MK. ΙΟ. 1 3-45

The judgment that the material of the longer text was probably coeval with 10.17—

34 is further supported by the fact that they fit together as parts of a larger section,

THE SECRET GOSPEL

10.13-45— a pericope designed to provide a textual basis for systematic teaching concerning baptism, teaching which followed the baptismal service point by point.

Accordingly, the pericope may have been read at the baptismal service preceding the pascha. This was first pointed out to me by Richardson in a letter (of January

13, 1961) which furnished not only the thesis, but also some important items of the evidence I shall use to support it. Though the bulk of the following argument is mine, I have profited so greatly by discussions with Richardson that it would be impossible to indicate exactly the limits of his contributions. This general acknowl edgment — with my thanks — must stand in place of many brackets and initials.15

The evidence for treating the whole passage, 10.13-45, as a baptismal pericope is as follows:

(a) Clement says the longer text was written els την των τελειουμενων χρησιν (1.22) and was read ττρος αυτούς μόνους τούς μυουμενους τα μεγάλα μυστήρια (II. 2). This proves its liturgical use in his church; and τά μεγάλα μυστήρια would most easily be referred to the pascha, the annual occasion for baptism. Clement sharply distinguishes ordin ary baptism — το λουτρόν — as the lowest stage of the Christian initiation, from τά μεγάλα μυστήρια — see the commentary on II. 2; but Clement's church may have practiced a second baptism by which the believer achieved true gnosis. This question must be postponed to Chapter Five; here it is enough to remark that such higher initiatory rites were common in second-century Christianity (Acts 8.14-17; 19. 1-7;

Irenaeus (Harvey, I.14.1 = Stieren, I.2i.if). [C.R. thinks a second baptism un likely; he would explain το λουτρόν as baptism administered alone and τά μεγάλα μυστήρια as the entire paschal ceremony, including baptism.]

(b) The liturgical use of this pericope in a most important service in Clement's church would account for (1) the fact that almost half of Clement's quotations of Mk. come from chs. 9 and 10 (above, p. 79, and Appendix D) ; (2) the influence of this pericope on the western text (above, p. 122) ; (3) the frequency with which it is echoed by the other parts of Mk. (above, pp. 1 39fT) ; and (4) the fact that Luke ends his "great insertion" (9.51-18.14) and comes back to the text of Mk. precisely at the beginning of this pericope (Mk. 10.13 = Lk. 18.15).

(c) That the service — the μεγάλα μυστήρια — with which Clement associated this material was (or included) some sort of baptism may be indicated by the scriptural reminiscences with which he justifies himself for quoting the passage to Theodore

(II.i6ff): ημείς δε υιοί φωτός εσμεν, πεφωτισμένοι τη εξ ύφους ανατολή του πνεύματος του κυρίου * ου δε το πνεύμα τού κυρίου, φησίν, εκεί ελευθερία , πάντα γάρ καθαρά τοΐς καθαρό ις.

The baptismal associations of these verses are familiar; the references in the com mentary, above, could easily be multiplied.

(d) The structure of canonical Mk. can be seen as parallel to that of the paschal service: chs. 1-8.30 can be taken as exoteric teaching for those catechumens not yet ready for baptism. Then, with the creed (Peter's confession) and the consequent transfiguration would begin the esoteric teaching for those about to be baptized (so 15. A similar interpretation of the longer text alone as a baptismal lection was suggested by G.L., but this suggestion rested on interpretations of detail which I do not think plausible and therefore have not presented.

THE SECRET GOSPEL

Riesenfeld, Tradition 162). Here the baptismal pericope would follow. After the baptism the story, like the service, proceeds to Jerusalem, the eucharist, passion, and resurrection, and closes with a hint of the disciplina arcani in 16.8. It must be admitted that this outline is obscured by numerous interpolations in Mk., but one could defend the proposition that it remains recognizable. Its recognition would confirm the hypothesis that 10.13-45 is a baptismal pericope: the section stands at the very place in Mk. where the general outline would require such a pericope.

(e) Finally, the details of Mk. 10.13-34 + the longer text + 35-45) and the order in which they occur, correspond to the content and order of the elements in the baptismal service. This I shall now demonstrate by a commentary (points i-xii) :

(i) The pericope begins with Mk. 10. 13-16, a pronouncement story of which the key verses are 14—15, άφετε τα τταιδία ερχεσθ at ττρός με, μη κωλύετε αυτά , των γάρ τοιουτων εστιν ή βασιλεία τον θεοΰ , (15) άμην λέγω ύμΐν , ος αν μη δεξηται την βασιλείαν του θεοΰ ώς παιδίον ού μη είσελθη εις αυτήν. 1 5 is omitted by Mt. and may not have stood in his text; Bultmann ( Geschichte 32), Lohmeyer, and Klostermann (both ad loc .) recognize it as a secondary addition. Mt. has a different version in 18.3 (the common supposition that this derives from Mk. — e.g., my Comments 45 — is gratuitous).

Another presumably underlies Jn. 3-3ff. Whatever the original sense of the saying,

Jn. is conclusive evidence of its interpretation to refer to baptism, and the same inter pretation probably explains its addition to the stories here and in Mt. 18 (cf. Mt.

11.25,29 άρατε τον ζυγόν = rVDbfr Vll? = δεχεσθαι την βασιλείαν — the essential act of the convert. This explains the apparent contradiction: only those who accept the kingdom can enter it, Dodd, Parables 34.) Cullmann, Baptism

72ff, has established the likelihood that there was an early baptismal formula τί κωλύει; ούδεν κωλύει. Of this the μη κωλύετε in the text may be an echo. Baptism as rebirth, and the childlikeness of candidates and recipients, are commonplaces of early

Christian literature; besides the passages cited above, see I Pet. i.igf; 2.2; Hermas,

Mandates II. 1 ; Similitudes IX. 29.1 ; Acta Thomae 132; Gospel of Thomas (Leipoldt) 22;

Clementine Homilies VII. 8; XI. 26; Sibylline Oracles VIII.3i3ff; Grant, Children 71.

Accordingly we need not enter the current controversy as to whether Mk. io.igff was written to justify infant baptism (so Jeremias, Kindertaufe; contra, Aland, Sauglingstaufe). Admittedly the passage was used in Tertullian's time to justify infant baptism

( De baptismo XVIII), and when infant baptism was provided for, infants were bap tized first (Hippolytus, Apostolic Tradition XXI. 4), as Mk. 10. 136? stands first in the Markan pericope. But the position of the passage would be justified even in a pericope dealing solely with adult baptism, since it states a sine qua non. This is why its parallel also stands first in the discussion of baptism in Jn. 3. This was almost seen by Grundmann, on Mk. 10. 17-31 : "One must reckon with the possibility that the pericopae concerning the blessing of the children and the rich young ruler, with the attached conversation with the disciples, had been put together in some pre-Markan tradition as answers to the question: How can one enter the kingdom of God?"

(ii) Having begun with the blessing of those who are like children (in what way need not concern us), the pericope in Mk. goes on to more specific requirements for baptism (verses 17-22): monotheism, observance of the Ten Commandments,

THE SECRET GOSPEL

renunciation of property.16 These are presented in answer to the question, "What shall I do to inherit eternal life?" Equivalent questions in Acts 16.30 and 2.37 lead directly to baptism; cf. also Acts 1 1.18, where baptism is την μετάνοιαν εις ζωήν (with a preceding reference to the κωλνειν formula, as here). In Lk. 10.25 the same question is answered by the two great commandments, which also appear in the Didache as the first point of the instruction to be recited before baptism (1. 2, cf. VII. 1). Cf.

Justin Martyr, First Apology 61.2, the preparation for baptism: όσοι άν ττεισθώσι καί

7 τιστευωσιν άληθη ταΰτα τά ύφ ' ημών διδασκόμενα και λεγάμενα είναι (that is, monotheism and the role of Jesus), και βιοΰν οϋτως δννασθαι ύπισχνώνται (according to the com mandments of Jesus) — these proceed to baptism. Baptismal catechesis presumably lies behind the similar summary of Christian teaching in Aristides, Apologia 15.3-9

(monotheism, commandments, sharing property with the poor). Again, in the passages behind which Boismard, Liturgie , has recognized a baptismal catechism (I Pet. i. 3-1 2; Titus 2. 12-14; I Jn. 3.1-11 — especially this last) appears the same sequence as in Mk. 10. 13-22: children of God, the holiness of God, the consequent obligation to keep the commandments and to share with the poor ( ποιεΐν δικαιοσύνην) .

On I Pet. 1.3-2 1 see also Windisch-Preisker, Briefe 157. Preisker likewise sees in this a prebaptismal διδαχή. Note also the sequence of themes in 17-21 : the holiness and fatherhood of God, the fear ( = obedience) of God, the worthlessness of silver and gold, the saving death and resurrection of Christ. This same sequence appears in Aik. 10.17-34. The evidence given by Preisker and Boismard seems to me to refute the contention of Robinson, Survey , that in NT times there was no prebaptismal instruction. Robinson's arguments are all based on silence; and the most impressive instances of silence, those in Acts, are not probative because most of them report baptisms resultant on miracles and therefore not to be taken as evidence of normal procedure.

(iii) In these passages used by Preisker and Boismard, however, the holiness of God is taken as an attribute to be realized also in the lives of his children. In Mk.

10.18 the goodness of God is sharply declared unique: τι με λεγεις αγαθόν ; ονδεις αγαθός εί μη εις 6 θεός. This declaration is clearly an insertion irrelevant to the latter half of Jesus' reply ( τάς εντολάς οΐδας, κ.τ.Α.) and to all the rest of the story. Only the latter half of Jesus' reply responds to the preceding question, and only this latter half is found in the independent version of the same story in Lk. 10.256 Moreover, the Didache (I.2) parallel to Lk., introducing the commandments as η οδός της ζωής, suggests a pre-Christian, Jewish origin for the present question and answer in its

Lucan form, and a variant of the same question and answer appears in many passages of rabbinic literature as an exegesis of Ps. 39. 1 3iT (Margulies, Wayyikra Rabbah , sec.

16.2). Accordingly, the insertion in the Markan form has to be explained. Why should an editor have inserted such a detail? To adapt the story to the requirements of baptismal catechesis. The catechesis required an initial proof text for monotheism, so a verbal detail ("good master") was seized on as an excuse to insert this irrelevant proof text. That 16. Walter, Analyse, has not persuaded me that the following account should be changed.

THE SECRET GOSPEL

ού8ευς αγαθός, ό θεός. ευ μη εις is a proof text is clear not only from its mnemonic form, but also from its background and its history in Christian exegesis. Its background is generally in acclamations of the εΐς θεός type (studied by Peterson, Εΐς), but more specifically in Philo, De mutatione nominum 7 · Moses, πάντα διά πάντων ερευνησ ας εζητει τον τρυπόθητον καί μόνον αγαθόν τηλαυγώς ί8εΐν, and therefore prayed to see God (Ex. 33.13). Even more striking is

De somniis 1. 148— 149, ταΐς μεν 8η των άκρως κεκαθαρμενων 8υανοί αυς άφοφητί μόνος άοράτως δ των όλων ήγεμών εμπερυπατεΐ . . . ταΐς 8ε των ετυ άπολουομενων, μηπω 8ε κατά τό παντελές εκνυφαμενων . . . αγγέλου, λόγου θείου, φαυ8ρννοντες αντάς τους καλοκάγαθίας

8όγμασυν. όσα 8ε εξουκίζεταυ κακών ουκητόρων στίφη, υνα εΐς δ αγαθός ευσουκίσηταυ,

8 ηλόν εστυ. σπον8αζε ούν, ώ φνχη, θεοΰ οίκος γενεσθαυ, ιερόν άγυον. The connections of this passage with the rhetoric of Christian baptism are so numerous that it may be thought additional evidence for the baptismal usage of the εΐς ό άγαθός/θεός phrase and consequently of the Markan pericope to which the phrase was added. As for the history of Mk. 10.18 in Christian exegesis, that shows the verse constantly used, as in this passage, to inculcate monotheism. It appears in Justin,

First Apology 16. 6f, as proof ώς . . . τός θεόν μόνον 8ε ΐ προσκννεΐν, and Dialogue ιοι.ι as a proof text against the Marcionites — Jesus was saved by the God of the OT.

Marcion, for his part, used it to prove that the true God, qua good, is superior to both Jesus and the demiurge (Hippolytus, Philosophumena VII.31). In Ptolemaeus it is proof of the nature of the τελευος θεός ... δ πατήρ, εξ ον τα πάντα (Epiphanius,

Panarion XXXIII. 7. 5^)· Irenaeus (Harvey, 1. 13.2 = Stieren, 1. 20. 2) reports that the Marcosians used it to the same purpose; so, too, the Naassenes in Hippolytus Phil osophumena V.7 (fob 30 verso). It was not only an important text in gnosticism (Grant,

Gnosis 4), but also a favorite of Clement (Osborn, Philosophy 65) — again, as a proof of a single, supreme God τον αγαθόν καί πρώτον καί μόνον ζωής αιωνίου ταμίαν, ην 6 υιός 8ί8ωσυν ημυν παρ' εκείνου λαβών (III. 164. iif — -again a baptismal theme).

(iv) The use of the ten commandments in baptismal teaching, as in Mk. 10.19, appears already in the Didache (II. 1-3) and probably in Pliny's epistle of the year 1 12

(X.96.7; Grant, Decalogue 1 if) ; it was anticipated by their use in connection with the shema' — the nearest thing in Israelite tradition to a confession of faith (Kuhn,

Phylakterien; Grant, Decalogue 1). [Stendahl reminds me that in Christian baptismal practice not the whole decalogue was used, but only the latter half, as in Mk. 10.19.

See his School 63.] (v) εμβλεφας αυτώ ηγάπησεν αυτόν (Mk. 10.2 1) is not paralleled in either Mt. or Lk. Either the later evangelists deleted it as improper (so Wellhausen EM, ad loc.), or they did not find it in their texts of Mk. Its connection with the rest of the Markan story is difficult to explain. Does it imply that because Jesus loved the young man he told him that in order to be saved he must sell his belongings and follow him? If Jesus had not loved him would he have concealed this secret and let him go to hell ? (So Lohmeyer!) Or does it imply that because Jesus loved him his fulfilment of the com mandments, which would otherwise have sufficed for salvation, became insufficient

THE SECRET GOSPEL

and this extra requirement (which he would not meet) was imposed on him by a special act of divine love? (So Lagrange!) Such "explanations" suggest (by their desperation) that the phrase here is intrusive. So does the fact that it is not in Mk.'s manner, when repeating himself, to transfer to one person what he says elsewhere of another. Such transferences are more common in the use of Markan material by the later evangelists (see above, p. 137). Accordingly, the phrase here is an inter polation and the longer text enables us to explain it. It was added here to identify this man with the youth of the following passage in the longer text, where the same phrase is used with a clear narrative function: the youth, looking at Jesus, loved him, and therefore begged to be with him, and therefore the two of them went to the youth's house. This makes sense. The same hand probably added ην γάρ εχων κτήματα πολλά to the longer text (III. 6) for the same purpose of identifying the two characters (and some copyist abbreviated it to rjv γάρ πλούσιος; see above, in the commentary, ad loc.).

The editor who made these interpolations (both in canonical Mk. and in the longer text) seems to have worked after Matthew and Luke, but worked so early that his addition has probably left no trace in the MS tradition of Mk. 10.21. (The omission of ηγάπησεν αυτόν και by ii, 1 5, and 579 is more likely censorship.) However, it is notable that Mt. 19.20 makes the unidentified rich man of Mk. a "youth" — νεανίσκος

— like the one in the following passage of the longer text, and thus effects the same identification as did the editor of Mk., but without the awkwardness of the intrusive phrase and the possibly objectionable attribution to Jesus of love for a man. Here we have perhaps another indication that Mt. knew the longer text (cf. the com mentary on προσεκυνησεν, 11.24, and on και προσελθών, III. i, above). The motive for identifying the two characters was the baptismal use of the pericope, for ηγάπησεν αυτόν in 10.21 looks forward to the resurrection and the nocturnal initiation (baptism) in the longer text. Baptism is the gift of love. So Eph. 5-25ff ό χριστός ηγάπησεν την εκκλησίαν καί εαυτόν παρεδωκεν ύπερ αυτής ΐνα αυτήν άγιάση καθαρίσας τώ λουτρώ τοΰ ϋδατος εν ρήματι. Apoc. 1.5, τω άγαπώντι ημάς καί λυσαντι ( λουσαντι , koine, Pal.gig.Vg.) ημάς εκ των αμαρτιών ημών εν τώ αΐματι αύτοΰ. Jn. 3.16 and 13. iff. (Is footwashing a Johannine second baptism?) (vi) The abandonment of property as a requirement for baptism was evidently a peculiarity of that church from which this material originally derived. The corollary and compensation of this primitive communism appear in verses 28-30: Whoever joins the group enjoys its common property and is a member of its common family. νυν εν τώ καιρώ τουτώ . . . μετά διωγμών is hardly explicable otherwise. So the ancient

Church understood the passage (Cramer; Theophylact) and so, generally, do modern commentators (cited by Lagrange, ad loc.). Goguel's objection (

Persecutions 275 n3) that community of goods was "never" realized is a. petitio principii', it apparently was realized in the community from which this text came (cf. Lohmeyer, 2 14-2 19;

Grundmann, ad loc.). The parallels discussed above, p. 165, if valid, would show that verse 28 (Peter's claim, "We have left all and followed") was already attached to the story in the earliest form of Mk. Verses 26-27 are clearly secondary (Bultmann,

Geschichte 21). They reflect the abandonment of the original requirement and explain that, in spite of the famous saying about the camel, salvation is possible even for those

THE SECRET GOSPEL

who keep their wealth. A similar modification appears in Mt. 19.21, where the un conditional requirement of Mk. is softened to a counsel of perfection : ει θε λεις τελειος είναι. Since the Markan requirement was thus soon abandoned, no survival of it is to be expected in the baptismal teaching of the later Church. But it is in place in a baptismal pericope, where later baptismal teaching substituted for it the requirement of indiscriminate charity: Didache 1. 5, παντι τω αίτοΰντί σε δίδου και μη άπαί τει.

(Further passages cited above, ii, end.) The rewards in this world are the consequences of church membership, effected by baptism (verses 28-30), "and, in the world to come, eternal life," made available by baptism.

(vii) Hereupon follows the prophecy of the passion and resurrection because, in the first place, it is the essential of the specifically Christian creed; and in the baptismal service, as here, this creed follows the preliminary instructions in monotheism, com mandments, and charity, which Christianity had in common with other forms of Judaism (cf. the "Two Ways," in Didache I— VI). Mark also attached the passion and resurrection prophecy to Peter's confession in eh. 8, of which it fills out the credal content (verses 29-31) : Son . . . suffered . . . dead . . . rose again. Like the creed, the passion prophecy is here presented as a μυστήριον (Theophylact, ad loc.)3 "esoteric teaching" (Bultmann, Geschichte 357). In the second place, Jesus' resurrection is the assurance of the efficacy of baptism and the reliability of the promise — in the pre ceding verses — of reward in the world to come (I Cor. 15.12-22). Finally, Jesus' passion, death, burial, and resurrection are, according to Paul, the essential content of baptism; the effect of baptism is to make the initiate participate in these (Rom.

6.3f). And this same use of baptism as the symbol of passion, death, and burial ap pears in the immediately following section of Mk. (10.38, δύνασθε ... το βάπτισμ a δ εγώ βαπτίζομαι βαπτισθηναι ;) . It is therefore plausible to suppose that the passion prophecy appears in this baptismal pericope not only as the assurance, but also as the explanation of the rite's efficacy.

(viii) Chrysostom, De quatriduano Lazaro , treats the Lazarus story as the immediate sequel and consequence of the prophecies of the passion, and so, too, it is here, in the longer text of Mk. Chrysostom sees in it Jesus' demonstration of his power to raise the dead, thus the antitype of his own resurrection. For the author of the longer text it was probably also the antitype of the two resurrections of the believer, both the final resurrection which will bring the life of the world to come, promised above — this interpretation appears in Irenaeus (Harvey, V. 13 = Stieren, V. 13) — and also the initial, baptismal resurrection of the sinner from the death of sin — this interpretation appears in Origen, for example, in Homilia in Jeremiam IX. 3 (ed. Klostermann,

GCS Origines , vol. 3, pp. 6yf) and in Hippolytus Commentary onjn. (ed. Bonwetsch and Achelis, GCS Hippolytus , vol. 1 .II, p. 216), where the interpretation of the story as a preparation for baptism is explicit: on the statement that Lazarus was sick, Hip polytus comments, ώ ασθένεια πυρετούς φυγών αποσοβούσα και ιδρώτας βαπτίσματος δροσίζουσα, και λαμπρας στολάς φυγής εξυφαίνουσα. Moreover, the interpretation of

baptism as resurrection from the dead seems to have been held already by the Carpocratians (below, p. 185) and is traceable through the hymn quoted in Eph. 5.14 back to the time and possibly to the affiliates of the Qumran sect; see Kuhn, Epheser-

THE SECRET GOSPEL

brief 342-345. Accordingly, the story is perfectly in place where the longer text puts it in this baptismal pericope. The baptismal and the literal interpretations are con joined in the second-century and perhaps Egyptian Epistula Apostolorum , 27(38), where

Jesus says, "And therefore, indeed , I have gone down to Abraham, Isaac, and Jacob, (the place of ) Lazarus , and have preached

(to the righteous andy the prophets to your fathers the prophets, and have brought them word that they might come out of that they might come out of the rest, which is beneath, the rest , which is beneath , and ascend to that which is ( abovey into the heavens; • · · and I have given them the right hand right (han)d on them . . . of the baptism of life and forgiveness and release, freeing from all evil, as I have also done for you, and from now on also for those who believe on me."17 Here the saviour's entering the tomb (underworld) and reaching his hand to the dead agree with the longer text against Jn. Similar interpretation and agreements appear in the Apocryphon of John (edd. Krause and Labib, i95ff, 25off) and in Methodius, De resurrectione 1. 23 end (ed. Bonwetsch, GCS, p. 248) : In saving the wicked from sin πρεπει 8e τω θεω άνοιγε lv τα μνημεία έκαστου καί εζάγειν εκ των μνημείων ημάς εζωοποιη μένους , ώσπερ 6 σωτηρ τον Λάζαρον ειλκυσεν εξω. This agrees with the longer text against Jn. in making

Jesus open the tomb himself and (probably) raise Lazarus by hand.

Thus the Lazarus story, in both its Johannine and Markan forms, was connected with baptismal resurrection. This perhaps explains why John prefaces his version of the story by a contrast of Jesus with the Baptist, continuing his polemic against the latter (i.26ff; 3.26ff; 4.1). In logoff he has Jesus go to the Baptist's original territory and there emphasizes the differences between them: καί πολλοί ηλθον προς αυτόν καί ελεγον ότι Ιωάννης μεν σημειον εποίησεν ουόεν , πάντα όε όσα είπεν ' Ιωάννης περί τουτου άληθη ην. This leads directly to the raising of Lazarus, Jesus' greatest σημειον (that is, miracle, as in Jn. 11.47; 1 2. 1 8 ; cf. Melito, On the Passion 12, line 38), but the intention may also be to contrast the miraculous resurrection effected by the Christian σημειον (sign

— baptism: Clement III. 138. 15; cf. circumcision, Rom. 4.1 1) with Johannite baptism which — John implies — had no such supernatural effect; a similar contrast is made in Acts 19. 1-7.

(ix) The baptismal concern which has thus far dominated the pericope makes it probable that the nocturnal initiation which follows the Lazarus story should be understood to be a baptism. This probability finds curious confirmation in a tradition reported from Ephraem Syrus by Dionysius bar Salibi, that Lazarus was raised in order to be baptized, and that after his baptism he was taken to Alexandria (where the longer text appears). Baumstark, Lazarusakten, from whose discussion I know the i74 17. My translation from Duensing's German. The italics are his and distinguish elements peculiar to the older, but imperfectly preserved, Coptic text, from the complete, but corrupt, Ethiopic. His revised translation (Hennecke-Schneemelcher I.141) presents no significant difference.

THE SECRET GOSPEL

passage, thinks the source must have been a very early 77 -ερίο8ος, πράξεις, or μαρτυριον

Λαζάρου (p. 2i i). Another trace of the same tradition perhaps appears in the Iohannis

Evangelium Apocryphum Arabice (tr. I. Galbiati, Milan, 1957), LIII.8, etc., which refers to "the disciple whom Jesus loved and whom he instructed in his mysteries"; cf.

Clement's description of the rich young ruler as υπο του κυρίου συντελειουμενος (11.22 1.

27). These passages may reflect knowledge of at least the content of the longer text, where the baptismal character of the initiation is indicated not only by the preced ing context, surveyed above, but also by the details of the rite : it is after six days, it is nocturnal, the prescribed costume is a sheet to be worn over the naked body, and the con tent is the mystery of the kingdom of God. Let us examine these details one by one.

"the six days," Richardson wrote to me, "are the six days of the Alexandrine paschal feast (Dionysius of Alexandria, Epistle to Basilides 1 end). Cf. the transfigura tion six days, Mk. 9.2, and the paschal six days of Jn. 12. 1. The earlier custom of two days is in Mk. 14.1, Didache VII. 4, and Hippolytus, Apostolic Tradition XX. 7;

XXIX. 2, and is referred to by Dionysius, loc. citT However, Dionysius wrote of the length of th & fast before the feast celebrating the resurrection, and his letter makes clear that even as to this there were variations of practice. Mk. 14.1, μετά 8υο ημέρας, supposes a three-day preparation period (see above, p. 163 n8). Thus the story of the anointing, which it places on the first day, accords with the preparatory washing of the catechumens which Hippolytus, Apostolic Tradition XX. 5, sets on Maundy Thurs day, to be followed by the fast on Friday and Saturday. And Hippolytus XX. 1-3 supposes that even before these three days there had been a preparatory period in which those set apart to be baptized had undergone examination, Gospel reading, and daily exorcism. A seven-day period of preparation for baptism is supposed in Acta Thomae 26 end. Jn. 12.1, which puts the anointing προ εξ ημερών του πάσχα, probably presupposes such a period, and the importance in pre-Markan tradition of a six-day period prior to the initiatory revelation is strongly suggested by the simi larities (set forth above, section 2) of Mk. 8.29-9.8; 10.20-32 plus the longer text; and 14.27-16.5. So it seems likely that the six days' preparation before baptism is even earlier than the two days' fast of the Didache. Both sevenand three-day periods of preparation are frequently required for magical operations; e.g., PGM III. 304;

V. 228; VII. 334; XIII. 1 15, 1 18, 674.

nocturnal: Hippolytus, Apostolic Tradition XXI. 1, sets baptism at cockcrow, fol lowing an all night vigil. This is repeated by the later Church orders (e.g., Apostolic

Constitutions V. 19.3) and seems to have been common usage. Baptism is nocturnal in Acta Petri et Andreae 21 end; Martyrium Matthaei 8; Acta Thomae 27; and probably

Clementine Homilies XIV. 1 . If the author of Acts 1 0.47 knew the geography of Palestine, he must have thought baptism nocturnal: Caesarea is 10 hours by horse from Joppa

(Baedeker, Palestine 239) so Peter would have arrived in the evening. Baptism is again nocturnal in Acts 16.33. The footwashing in Jn. 13 (a variant of baptism) is nocturnal.

Nicodemus, in Jn. 3, comes to Jesus by night and receives instruction concerning

baptism as the means of entering the kingdom of God (3*3?5? 1 3^ ) · the sheet over the naked body: Nudity in baptism is prescribed by Hippolytus

Apostolic Tradition XXI. 3, 5,1 1, and was required by the Pharisees in proselyte baptism

THE SECRET GOSPEL

as well as in immersions for purification ( Mikwaot VIII end and IX; B. Tebamot 47b). Particularly close to the longer text, where the prescribed garment is a σινδών, is Acta Thomae 12 1, where the apostle has to baptize a young lady: εκε λευσεν τη τροφω αυτής άποδυειν αυτήν καί σινδόνα αυτήν 7 τεριζώσαι. Undressing for baptism ap pears also in Acta Barnabae 12 and Martyrium Matthaei 27. In the Johannine foot washing (which Aphraates, Demonstratio XII. 10, interpreted as a baptism) Jesus is naked except for a towel (Λ ε'ντιον, 13.4)5 which the Syriac versions describe as a σινδών

(kjotod) . Is it by chance that the beggar in Mk. 10.50 (immediately after the second passage of the longer text) throws away his himation when he comes to Jesus to be cured? The gesture was understood as symbolic by the commentator in Cramer, ad loc. and may have been so understood by the author. That naked baptism was already customary in Paul's time is shown by his allegorizing the undressing for it and dressing after it: Col. 2.1 if, εν τη άπεκδυσει του σώματος της σαρκός , iv τη περιτομή τοΰ χρισ τού ... εν τω βαπτίσματι', Gal. 3 * 2 7> οοοι γάρ εις χριστόν εβαπτίσθητε χριστόν ενεδυσασθε; further, I Cor. 1 5*33^ (as interpreted by Odes of Solomon 15.8, see Lietzmann,

Korinther, ad loc.) and II Cor. 5.2. Similar allegorization (not apparently dependent on Paul) appears in the Gospel of Thomas (Leipoldt) 37; the Gospel according to the Egyptians (Clement, II.238.24ff) ; and P. Oxy. 655 end; and the same general theme recurs in the Gospel according to Philip (Schenke, 77, 101) ; Clementine Homilies VI II. 2 2. 4—

23.1; Hippolytus, Philosophumena V. 19 end (the Sethians) ; Irenaeus (Harvey, Justin, Dialogue 116; Clement III.131.25ff (

Excerpta ex Theodoto)-, 140. 6ff; I43.24ff [Eclogae) ; Apostolic Constitutions VI. 6; and many later Christian documents. This early dissemination of the theme argues an early origin.

The indicated baptismal practice (nudity and σινδών) lent itself particularly to Pauline exegesis because (as Lieberman pointed out to me) the initiatory σινδών

(which LXX regularly uses to translate po) was also the regular burial garment; cf. J. KiVayim IX. 4 (32b) = J. Ketubot XII. 3 (35a) ΌΊ ηηρί "ΤΠΝ ]Ή02, "Rabbi

<( Judah, the Patriarch) was buried in a single σινδών " (and no other garment);

J. Terumot VIII. 10 (46b, inf.), 1ΓΗ02 Γ)?3Π "pD'', "Let the dead man be wrapped in his σινδών " (that is, let the man be abandoned to his fate — a current saying).

Accordingly, Jesus had been buried in a σινδών (Mk. 15.46 and parallels). Paul's interpretation of baptism as death and burial with Jesus suggests that the σινδών over the naked body was the customary costume as early as Paul's time.

The σινδών was not a "proper" garment, though it was worn occasionally. Crates, the cynic, when called down by the Athenian police for going about in a σινδών, offered to show them the great philosopher Theophrastus similarly attired. When they found this incredible, he took them to a barber shop and showed them Theo phrastus in a σινδών having his hair cut (Diogenes Laertius, VI. 90).

Accordingly, since the costume specified in the longer text is unusual and is as sociated with baptism, it can be used as an indication of baptism. This explains why the phrase of the longer text, περιβεβλημενος σινδόνα επί γυμνού, recurs verbatim in Mk. 14.51 (except that W fam 1 c k Sy.s-Cop.sa· have discreetly omitted επί γυμνού,

14. 1-4 = Stieren, 1. 21. 1-5 — the Valentinians) ; Acta Thomae 132 (text P only); while Θ fam 13. 543-565.Sy.pAeth. have accidentally replaced it by the γυμνός of the

THE SECRET GOSPEL

following verse). In 14.51, too, the subject is νεανίσκος τις — the young man who was with Jesus at the time of his arrest and who, on being seized, fled naked (an episode both Matthew and Luke chose to omit). As suggested in the commentary on III. 8, above, the recurrence of the περιβεβλημενος phrase is to be explained as that of a fixed formula, probably a baptismal rubric. Mark's fondness for repetition as a means of cross-reference was noted above (p. 136) ; for example, his initial identification of the Baptist as Elijah merely by repeating ζώνην δερμ ατίνην περί την οσφύν αυτού. And fixed formulas, especially those connected with baptism, frequently recur verbatim or are referred to by partial quotation in the NT and early Christian literature, for example, το βάπτισμα Ίωάννον, Mk. 11.30 and parallels; Lk. 7.29; Acts 1.22; 18.25;

19.3; (εις) άφεσιν αμαρτιών Mk. i. 4 and parallels; Mt. 26.28; Lk. 1.77; 24.47; Acts

2.38; 5.3 1? 10.43; 1 3*38i 26.18; Col. 1. 14; εις τδ δνομα τοΰ κυρίου ' Ιησού, Acts 8.16;

Ι9·55 cf. 2.38 (επϊ)\ 10.48 (εν) ; I Cor. I.i3f, etc.; τί κωλύει ; ουδέ v κωλύει , Cullmann,

Baptism 72ff (to his evidence add Clementine Homilies XIII. 1 1.1, where Peter's laughter is better explained if the previous speaker has unwittingly quoted a part of the formula of the rite). Further, Mk. 14.51 is echoed in Mk. 16.5, νεανίσκον . . . περιβεβλημενον στολήν λευκήν , and again in the Gospel of Peter 55 (13)3 νεανίσκον . . . περιβεβλημενον στολήν λαμπροτάτην', and the recurrent use of περιβεβλη μένους στολάς λεύκάς and cognate phrases to describe the saints in the Apocalypse (3.5,18; 4.4; 7.9,13; 19.8) indicates that the expression and the white garment (σινδών = στολή λευκή) had some special significance in the early Church. From the longer text and the parallels above, we can conclude that the garment was the baptismal and burial garment.

Thus the repetition in Mk. 14.51 (recognized by Bultmann as primitive tradition, by contrast with the "completely legendary" agony in 32-42; Geschichte 288ff) is an explanation, from an early Markan stratum, of what the young man was doing there

(a problem for which no plausible solution has hitherto been suggested). The reader who had already read the longer text would realize that this youth, too, had come to be baptized. (On the stylistic difference between the story of the arrest and that of the agony, see Wohleb, Beobachtungen 190; the objections of Zerwick, Untersuchungen

4f are naive.) However, a further question has to be answered: Why, then, is the same phrase used for the angels at the resurrection and the saints in the Apocalypse ? Because the burial garment is also the resurrection garment. This is stated explicitly by the first two of the rabbinic passages cited above — J. KiV ayim IX. 4 (32b) = J. Ketubot XII. 3

(35a) : "A man is raised in the same clothes in which he is buried." (Such is, at least, the majority opinion). Accordingly, the baptismal shroud is also the robe of the saints and of the saviour, and hence of the caelicoli generally. This explains the stories in Acta Thomae 27, Actus Petri cum Simone 5 end, and Acta Barnabae 3, where baptism is followed by appearance of a young man in a white garment and a blaze of light: these are the angels whose appearance announces the initiate's resurrection by baptism. Finally, this conclusion is confirmed by and itself confirms the observations

of Goodenough as to the symbolic importance of white robed figures in Jewish art

(for example, Kraeling, Dura pi. LII, LIII, LXI, LXII, LXIII, etc.). Goodenough's point of departure had been the obvious importance of such figures in Philo and in

THE SECRET GOSPEL

early Christian art (Goodenough, I.28ff and the indices, under Robe ; further, IX.131168, where the argument is pushed to absurdity but not thereby wholly invalidated). the mystery of the kingdom of god: This phrase, too, recurs in Mk., at 4.1 1.

For the variants there and those of the Matthaean and Lucan parallels, see the com mentary above, on III. 10.

In Mt. and Lk. (as Lk. now stands) the plural "mysteries" which the disciples are given "to know" are the teachings of the Church — especially, in this instance, its interpretations of the parables. This misinterpretation was suggested by the context into which the logion was inserted in Mk. 4. 1 1 ; it would also have been suggested by the use of έδιδα σκεν in the longer text, if Matthew and Luke had read it.

That Mk. 4.1 if is an old logion, not an invention by the compiler of Mk., was shown by Jeremias ( Gleichnisse 7-12, partly anticipated by Lohmeyer, ad loc. and Manson, Teaching 77) and has since been widely accepted (Taylor; Cranfield;

Grundmann; Gnilka, Verstockung 23; Boobyer's attempt in Redaction to defend the unity of the passage is not convincing). The compiler of Mk. knew or represented the logion as addressed to Jesus' circle (ol π ερι αυτόν) ; a later glossator has added συν τοΐς δώδεκα (4. io; Bultmann, Geschichte 71). The compiler also took εν παραβολαΐς τα πάντα γίνεται to mean, "I teach only in parables," (cf. Mk. 4.33k), but Jeremias has argued that, once the logion is separated from its present context, this meaning is not necessary; he proposes to translate "ist alles ratselvoll" — "everything is puzzling," and this, too, has been widely accepted. If it be accepted, there is no need to refer τά πάντα exclusively to Jesus' verbal teaching (and such a reference would not be expected even in rabbinic Judaism, where students customarily learned from their masters' actions as well as from their words — e.g., B. cErubin 64b). τά πάντα may, therefore, mean "everything I do and say," or even "everything God does." Accord ingly, one cannot define from τά πάντα the meaning of the μυστήριον to which it might or might not be sharply antithetical.

The μυστήριον, therefore, must be defined — if at all — from the general sense of the verse and the similar usages elsewhere in the NT and in related works. In the first place, it is clear that in Mk. 4. 1 1 the μυστήριον cannot be the explanation that follows in 4.14-20, for it is something which already "has been given" in the past, something which Jesus' intimates have already received and which makes it possible for Jesus to give them now, as a further gift, the explanation of the parable. That they had already received the μυστήριον was what distinguished them from τοΐς όξω, which presumably meant for Mark what it means in I Cor. 5.12; Col. 4.5; I Thess. 4.12; that is, those outside the Church (cf. the parallels from philosophic schools cited by Bauer, Wb., εξω, β). Now that "mystery" which "was given" to members of the Church, which distinguished them from nonmembers, and which enabled them to be given the secret teachings of the Church, was baptism.

The word μυστήριον in Eph. 5.32 probably refers to the spiritual union effected by baptism. The argument in verses 25-32 runs as follows: Christ loved the Church, washed it with water and the word, hallowed it, placed it beside him in glory, and made it his own body, therefore he feeds and cherishes it; men make their wives their own bodies , therefore they should love, feed, and cherish them. The first half of this argument

THE SECRET GOSPEL

reflects the Pauline doctrine of baptism (including the enthronement with Christ; cf. Col. 2.12-3.4). Of this doctrine the words italicized at the beginning and the end were relevant to the author's purpose; the middle terms (washing, hallowing, and enthronement) were not. Why, then, did he include them ? Presumably because they were fixed parts of the doctrine of baptism. Therefore when he concludes to μυστήριον τούτο μεγα εστίν , εγώ δε λέγω εις Χρίστον καί εις τήν εκκλησίαν, the "mystery" is presumably the spiritual union effected by baptism and thence the rite itself which makes the Church the body of Christ by making Christ's spirit live in the members.

To this mystery the writer compares the spiritual union effected by physical inter course in marriage, and he finds a reference to both these mysteries in Gen. 2.24.

The same use of "mystery" to refer to baptism is found in I Cor. 2.6f σοφίαν δε λαλοϋμεν εν τοις τελείοις . . . λαλοϋμεν θεόν σοφίαν εν μνστηρίω, where it is utterly implausible to neglect the parallelism and to separate the initiated ( τελείοις ) from the initiation (μνστηρίω) . Allo's objection (I Cor., ad loc .) that the Corinthian converts were not τελειοι neglects the distinction between potential and actual salvation, which is basic to Paul's thought with its constant alternation between deity and depravity

(Rom. 7-8, II Cor. 4, etc.) Baird's attempt to explain away the terminology ( Mature ) is built on the false antithesis: either "a special and esoteric doctrine reserved for . . . the initiated" or "the Gospel, 'the word of the Cross.'" It never occurred to him that, to the ancients, baptism was an initiation— a τελετή, as Lucian said (Peregrinus

1 1). Anticipating arguments to be presented later, it may be said that a further reason for understanding the "mystery" in I Cor. 2.7 to be baptism lies in the fact that the "wisdom of God" revealed in it involves the secret of Christ's descent in disguise and his assumption of the body from the cosmic powers, for the purpose of subjugating them : σοφίαν . . . ήν ούδείς των αρχόντων τον αίώνος τούτον εγνωκεν· εί γαρ εγνωσαν, ούκ αν τόν κύριον τής δόξης εσταυρωσαν', cf. Ascension of Isaiah 10— 11. This secret of descent in disguise underlies the Pauline interpretation of baptism in Col. 2.15, where the second half of the process, the stripping off in the ascent, is referred to. And the conclusion of the process in I Cor. 2.9, as in Col. 2.12-3.4, is the participation of the baptized in Christ's resurrection, ascension, and session in glory, a οφθαλμός ονκ είδεν καί ούς ονκ ήκονσεν . . . όσα ήτοίμασεν ό θεός τοις άγαπώσιν αυτόν. (Note again, as in the longer text of Mk., the connection of baptism with αγάπη; cf. above, section v, end). Finally, the "wisdom of God" (I Cor. 2.7) given in the mystery is Christ

(1.24,30) and Christ is given in baptism, therefore the mystery is baptism. This wisdom is given only by the spirit (2.1 1) and the gift of the spirit is the function of baptism (Acts 2.38 and often).

Since the "mystery" of I Cor. 2.7 is baptism, it is not unlikely that in I Cor. 4.1-6, where Paul speaks of himself, Kephas, and Apollos as οικονόμους μυστηρίων θεοΰ, he refers to their function not only in general, as agents of the grand secret strategy of God, but also in particular, as administrators of the mystery rites of baptism and the eucharist, and thus of the salvation effected by these rites. The same sense can be

borne by the interpretation of this passage in Eph. 3.9 and of this in Ignatius, Trail.

2.3, where the contrast between the "mysteries" and "food and drink is that of I Cor. 11 between the eucharist and ordinary eating and drinking; the reference is

THE SECRET GOSPEL

particularly to 11.22. (Bauer's reference, Ig ., ad loc., to Acts 6.2 is not justified by the text.) All the above argument on μυστήριον has run counter to the common dogma that μυστήριον in the NT always has the sense of n or *710 and that these always mean "secret," never "secret rite"; so Bornkamm, μυεω ; Nock, Mysterious Alio,

I Cor., on 5.7; Klostermann, Taylor, and Cranfield on Mk. 4. 1 1 ; etc. But this dogma is false.

Paul sometimes did use μυστήριον to mean "secret," but there are a number of instances where he clearly did not: the "mystery of iniquity," that is, "the unlawful magic," in II Thess. 2.7 is not a secret, but a process — a secret process, no doubt, but the essential thing is not the secrecy but the process, which is already "working" to bring the coming of the evil one. So, too, μυστήριον in the magical papyri means not only "charm," but the whole magical ceremony and its consequences ( PGM I.131;

IV.476; etc.; contrast Rigaux, Thess., on 2.7). Similarly in Col. i.26f the mystery is not a secret but a process, χριστός εν ύμΐν, the indwelling and working of Christ in the baptized; cf. PGM I.i28ff, where το μεγα τούτο μυστήριον means not only the whole magical ceremony by which one receives ό κύριος του άερος (cf. Eph. 2.2) as an indwelling deity, but also all the consequences which follow from the reception: the continued possession and service of the deity, the status of an initiate, μα[κάρι]ε μυστα της ίεράς μαγείας. That Paul should have used μυστήριον to describe the process and consequences of Christ's indwelling in Christian initiates was probably a reflec tion of his use of the same term for the rite of initiation, baptism. As in PGM, the mystery produced the association of man and god, and the term was carried over loosely to its consequences. Nock's notion that Paul was not aware of the connection between μυεω, μύστης, and μυστήριον ( Mysterion ni) seems to me incredible. That Paul was thinking of baptism in Col. i.26fis indicated by the immediately preceding verses where he speaks of the body of Christ, the Church (25), ης εγενόμην εγώ διάκονος κατά την οικονομίαν του θεού την δοθεϊσάν μοι εις υμάς πληρώσαι τον λόγον του θεοΰ, το μυστήριον, etc. This is the vocabulary found in I Cor. 4. 1 where Paul is speak ing of his work as οίκόνομος of the mysteries ; and the concept of the Church as the body of Christ is that remarked above in connection with the references to baptism, since by baptism this embodiment was effected. Thus, the notion that Paul always uses μυστήριον to mean "secret" is false.

Further, the notion that when Paul does use μυστήριον to mean "secret" he can not at the same time use it to mean "secret process" or "rite," is also false. This has been shown by the examples above. For some consequences of this false antithesis, see above, in the comment on μυστήρια in the body of the letter (II. 2).

Finally, the notion that π and "no always refer to secrets and never to secret rites is also false. *770 in particular is not infrequently taken in rabbinic literature as a reference to the rite of circumcision; for example, Tanhuma, Hayye Sarah 4, where

Prov. 31.24, nrWtf |*HOj is glossed (by punning on "770) with the words nV^n 7T r'XTV Π7ΓΡ *770 "this is circumcision, of which it is said, 'The mystery ("770) of the Ford is given to those who fear him'" — Ps. 25.14. (A contributory element in the exegesis may have been the fact that ]*ho was the initiation garment, the σινδών; see

THE SECRET GOSPEL

above). More important is the fact that both no and n appear in Hekalot Rabbati (27.1 ;

28.3 ; 29. 1 ,2,4 ; etc.) as descriptions of the magical technique by which one is enabled to ascend through the heavens and be seated in the throne of God. Though the present form of this textis late, Scholem, Gnosticism , has shown that speculations on the subject of the ascent go back to a syncretistic Judaism of the early first century a. d. at the latest; and his argument is now confirmed by the appearance of their peculiar angelology in the Qumran texts, 3Q, 7.5 = Discoveries III. 99. Further, I have shown ( Observa ions end) that the account of the ascent in the Hekalot and that in the so-called

"Mithras Liturgy" ( PGM IV.475ff) go back to a common source of tradition, if not of writing. The "Mithras Liturgy" describes itself as μυστήρια (476), and means by this "initiation," for it prescribes that one may have a συνμύστης (line 732). The first reference to practice of a technique for ascent to the heavens may be that in Col. 2.18: "the angels whom he saw when going in." Into the heavens? (Cf. Observa tions 156). And ascent to the heavens and session with Christ on the right hand of God are described by Paul in the same context as the potential climax of the conse quences of Christian baptism (Col. 3. iff). So the supposition that n and/or no could have been used by Paul to refer to the rite of baptism as a "mystery" is not unsupported.

Judaism was often considered a "mystery" religion (e.g., by Plutarch, Quaestiones

Convivales IV. 6: Jewish doctrines and ceremones are τά r Εβραίων απόρρητα and the unknown Jewish god is identified as Dionysus because of parallels between the Jewish and the Eleusinian mysteries). Paul's Jewish contemporary, Philo, frequently de scribed the doctrines and ceremonies of Judaism as "mysteries" (Goodenough,

Symbols VI. 206-2 16; Wolfson, Philo I.43). And the rabbis took over the word μυστηρίου with the full range of its Greco-Roman meanings.

On the one hand they used it to mean "spell" or "magic" and identified "the mystery of Israel," bttf pIDD^, as the secret name of Yahweh. It was by pronouncing this that Moses killed the Egyptian: Wayyikra Rabbah 32.4 end, and Margulies' note ad loc. As in Greek, the word can mean not only the essential charm, but the whole magical praxis in which it is used; thus the Shunnamite says to Elisha, rrnm Vw pi»o»n nnx Ti&y ted» ηκ p ^ nna n*?nn& crnVK pi&o&n may

,ΊΠΙΧ "You practiced the mysteries of God when you gave me a son, to begin with, so now practice the mysteries of God and raise him from the dead." (

Shemot Rabbah 19.1.

This passage is of particular interest here because it shows these mysteries were con ceived as means of resurrection, as was baptism.)

Besides this magical usage, the rabbis also took over, as Philo did, the usage of μυστήριον for "mystery initiation." They used it in this sense to refer to the initiatory rite of Judaism, circumcision. For this they found Biblical basis in Ps. 25.14,

□ΙΓΎΙΠ1? ΊΓΡΊΠΊ PNT1? ΠΊΓΡ Y10, which LXX had translated κραταίωμα κύριος των φοβούμενων αυτόν' και η διαθηκη αύτοΰ δηλώσαι αύτοΐς , but Aquila translated απόρρητον (Theodotion and Quinta , μυστήριον) κυρίου τοΐς φοβουμενοις αυτόν' καί την συνθήκην αύτοΰ γνωρίσει (Theodotion, δηλώσει) αύτοΐς, thus rendering Τ10 3-S

"mystery" and equating the mystery and the covenant. Compare Is. 24.16, where ·»*? "'V Ή was omitted by LXX but translated by Symmachus and Theodotion

THE SECRET GOSPEL

το μυστήριόν μου εμοί, το μυστήριόν μου εμοί — a translation also reflected by an agraphon attributed to Jesus in Clement, the Clementine Homilies , and many later authors (see the commentary on 1. 12): τά μυστήρια εμοι και τοΐς υίοΐς του οϊκου μου.

As observed in the commentary, this agraphon is introduced in the Clementina as if it came from Mk., but is interpreted, by analogy with the Matthaean and Lucan interpretations of Mk. 4. 1 1 , as referring to teachings. Ps. 25. 1 4, however, is consistently taken in rabbinic literature as referring to circumcision. Thus Tanhuma on Gen. 17.2

{Lek, 19) comments: ΠΓΚ ♦□17*,Τ)Πι7 "ΙΓΡΊΠΊ VXI'b 'Π *710 Π'ΊΡΐ jrm TO ΤΡΊΟ ΠΙΠΚΊ

•omoxb hVt Vo? ριοοο π-ορπ nV^n it •vntV nV'w το νίπ — "'And I shall establish my covenant between us.' This is what is referred to by the verse, 'The mystery of Yahweh is for those who fear him and so is his covenant, to give them knowledge' (Ps. 25.14). What is this 'mystery' which he revealed 'to those who fear him'? This is the rite of circumcision. For the Holy One, blessed be

He, did not reveal the mystery rite of circumcision to any save Abraham." This exegesis occurs as part of a homily beginning with the verse ΠΤΠ ΓΓΠΊ ''ID1? "jbnnn

(Gen. 1 7.1), which is taken to mean, "Do as I tell you and you shall be perfect." Here LXX and Philo rendered DTD by άμεμπτος, which Aquila corrected to τελειος — a term cognate to the technical terms "to initiate" (τελεω),

"initiation" (τελετή), etc. τελειος is also the term used by Jesus in urging his followers to follow his law rather than the traditional commandments; he concludes (Mt. 5.48), εσεσθε ουν υμείς τελειοι ώς ό Πατήρ υμών ό ουράνιος τελείάς εστιν ; cf. Mt. 19*21, where, in the conversation with the rich young man of Mk. 10, Matthew's Jesus contrasts a man who merely keeps the Mosiac law with one who is τελειος.

The difference between LXX and Philo on the one hand and Aquila, Symmachus, Theodotion, and Mt. on the other in the translations of Gen. 17.1 and Ps. 25.14 and Is. 24.16 (noted above) indicates that the interpretation of circumcision as a mystery is to be dated in the first or early second century. The sermon in Tanhuma goes on to declare this mystery the necessary means to happiness in the afterlife (as were, for example, the Eleusinian mysteries, and baptism). This exegesis was standard. It recurs in the parallel version of the Tanhuma (ed. Buber, Lek 20-27), which places even more emphasis on the notion that by circumcision the initiate becomes τελειος

(here 20 and 25), takes Abraham as prefiguring the Messiah (22), and ex plains that the mystery was given him in order that he might be made like God (23 end; cf. Mt. 5.48, above) and that its performance prevents the world's returning to chaos (24) — a notion derived from the mysteries of ancient Mesopotamia (and a reminder that this development need not be conceived as wholly the result of Greek influence). We have already seen this same exegetic complex referred to en passant in Tanhuma ( Sarah 4) apropos of a mention of the σινδών which was probably the costume for the rite (■»*» to *7?2N2ttf IT). A corrupt version was introduced into most MSS of Bereshit Rabbah 49.2; see Theodore's notes there. Finally, the theme is further developed in Shemot Rabbah 19.5-7. After emphasizing again that to have been circumcised is necessary and sufficient (save to heretics and the very wicked) for salvation, the text goes on to treat it as the precondition for participation in the sacred, secret, passover meal (6). Only one speaker is named — the Galliean, R. Simon

THE SECRET GOSPEL

ben Halafta, of the beginning of the third century — but how much of the tradition can be attributed to him is not clear. The text commented upon is Ex. i2.43ff,

•13 VdN'' kV Vm in VdK*· nV Ί33 ]3 Vd nosn ηρπ ΠΝΤ . . . Particularly striking is the comment, "It is like a king who made a dinner for his friends (ϊΟΠΊΝ1? = τοΐς άγαπώσιν αυτόν = τοΐς TeXeioig . . . ev μύστη ρίω, I Cor. 2.6—9; see above, page 1 79).

The king said, ' If there is anyone, of all those invited, on whom my seal does not appear, he is not to be admitted here.' <(Cf. Mt. 22.1-14. Circumcision is a "seal" in Rom.

4. 1 1 ; baptism or chrism — probably — in Apoc. 7.2 ; 9.4. All these use σφραγίς, on which, and on its background in the mysteries, see Bauer, Wb., s.v., and the literature there.)

So God made a feast <Te., the passover) . . . because he was delivering them from evil <ΓΠ^Π, cf. Mt. 6.13; Gal. 1.4). He said to them, 'If the seal of Abraham is not in your flesh you shall not taste of it' <(Lk. 14.24, also Didache IX. 5, none but the baptized may eat the eucharist) ... (7) 'No alien shall eat of it' . . . The Holy One, blessed be He, said to them, ' Let no other people participate in it, and let them not know the mystery <(]Ή1βΟ&, accepting the emendation of Krauss, Lehnworter, s.v.y, but let it be for you alone."' The text goes on to describe the Passover in this world as an anticipation of the feast of the blessed in the world to come.

In sum, then, the rabbis often spoke of the initiatory ceremony of Judaism as a "mystery" which had been "given" to them. They thought this initiation prere quisite for participation in the sacred meal of the cult in this world, and for the life of the blessed — a glorified form of that sacred meal — in the world to come. Thus the clear evidence of the rabbinic material confirms the preceding interpretation of Paul. μυστήριον in Jewish usage regularly refers to the initiatory rite of circumcision, as in Paul's usage it refers to the initiatory rite of baptism which Paul equated with circumcision: Col. 2.1 I ; τη πβριτομη του χριστού . . . τω βατττίσματι.

Given the Pauline and the rabbinic usage, it is not surprising to find μυστήριον used as a technical expression for baptism in the community from which Mk. drew his material. Christian baptism is "the mystery of the kingdom of God" because it en ables those to whom it is given to enter the kingdom — so Jn. 3, Nicodemus' nocturnal visit to Jesus (of which the similarity in content to the initiation in the longer text is no less striking than the total independence of the two passages). Therefore those to whom Christian baptism has been given (Mk. 4.1 1) are in the kingdom, as opposed to "those outside" (ibid.) like the Baptist (Lk. 7.28; Mt. 11.11; Acts 19. 1-7), who have been born only of women, not of water and the spirit (Jn. 3.3-5).

The one difficulty in the way of a baptismal interpretation of "the mystery of the kingdom of God" in the longer text is the word έδιδα σκε. In the light of the evidence reviewed above, I think this a corruption of an original 08ωκ€ν. The corruption was probably made by a copyist influenced by the misinterpretation of Mk. 4. 1 1 in Mt.

13. 1 1 and Lk. 8.10, where "the mystery" becomes "the mysteries" — that is, the following interpretations of the parables, which "it is given" the disciples "to know."

That misinterpretation, in turn, probably resulted from the rapid growth of Chris

tianity, which made baptism an experience common to all members of a large group including many not completely saved; baptism therefore declined in prestige and there was a corresponding growth of secret teachings which professed to reveal

THE SECRET GOSPEL

something more. This is the development reflected by the Matthaean and Lucan emendations of canonical Mk. and by the corruption of the longer text.

Apart from this, the details of the nocturnal initiation (the six days' preparation, the coming by night, the sheet and the nudity, and the mystery of the kingdom) all accord with the interpretation required by the clearly baptismal character of the preceding pericope. The rite was a form of baptism.

(f) This baptismal interpretation explains both the inclusion in the longer text of the story of the baptism and its omission from the canonical text if that was inten ded, as Clement says it was, for reading by the unbaptized.

The inclusion at the place where it stands is exactly in accord with the require ments of the pericope as a reading for the baptismal service. We have had the prere quisites — become as children, believe in the one good God, obey the commandments, forsake all, and join the Church — and the peculiarly Christian creed, the death and resurrection of Jesus, the Son of Man, which the initiates are now to share. We have had the evidence of resurrection and the antitype of baptism — the raising of Lazarus

(the Gospel for the service, her e following the creed— evidently because it belonged to the secret teaching of the church — see above on τούς μυουμενους in II. 2). Now we come to the baptism itself, and here, as in the eucharist, the essential is not a prayer , but a story of what Jesus once did: eVeV αξεν αύτω . . . καί . . . ερχεται . . . καί εμεινε

. . . εΒωκεν γαρ αύτω ... το μυστήριον της βασιλείας του θεού (III. 7—9)· And with this the story ends as abruptly as the Markan account of the institution of the eucharist: εκεΐθεν δε άναστας επεστρεφεν ( III. IO— Ii); cf. Mk. 14.26, ύμνησα ντες εξήλθον.

That this story should have been omitted in revision of the Gospel for exoteric use is in accord with John's omission of both Jesus' baptism and the eucharist (so Dodd,

Fourth Gospel 309-310). It is remarked in Windisch-Preisker {Brief e 157) that in I Peter, between 1.2 1 and 1.22, " Der Taufakt selbst ist aus Arkandisziplin fortgelassen."

Schille, Tauflehre 35, remarks a similar omission of advanced eschatological teaching in Barnabas 17.2: ου μη νοησητε διά το εν παραβολαΐς κεΐσθ αι; cf. Mk. 4·11· Preisker,

Boismard, and Schille represent I Peter and Barnabas as baptismal texts because of their outlines (above, page 1 70), and we have shown the extensive agreement of these outlines with that of Mk. 10. Therefore the omission in Barnabas 17 may help to explain the suppression in Mk. 10. Baptism, as the secret way of entering the kingdom

(Jn. 3.5) was acutely eschatological. The most important secret about the kingdom was how to get in. Perhaps this was why, in canonical Mk., baptism was suppressed while the eucharist was not (cf. Moule, Intention 17 if). This question, however, would lead to that of the structure and history of canonical Mk., which must indeed be re studied in the light of the longer text, but not here.

Returning to the role of the baptism story in the baptismal pericope, it should be noted that the liturgical interpretation of the whole pericope explains the seeming difficulty that the resurrection of the youth (the symbol of his baptism) precedes, by

six days, his baptism. One might either compare Apuleius, Golden Ass XI. 1 3-25 (where

Lucius' restoration to human form precedes his initiation, of which it is the symbol), or quote the Pharisaic ruling on circumcision: "Separation from a foreskin is like separation from a tomb." ηηρπ p uniBD p ΕΓΠΒΠ (

Pesahim

THE SECRET GOSPEL

VIII.8; 'Eduyot V.2), which puts a seven-day purification period (Num. 19.16) between circumcision and participation in the passover meal. But the Pharisaic ruling is not closely parallel, since the ceremonies it affects are not related as symbol and reality. And in Apuleius the duplication is due to the shift from allegory to auto biography: having allegorized his uninitiated self as an ass, he had to restore himself to human form (that is, undergo allegoric initiation) before he could describe his actual initiation as a man. In the longer text of Mk., however, the resurrection story

(allegoric initiation) anticipates the story of the actual baptism because the whole pericope is designed to correspond with a service in which the Gospel anticipates the actual ceremony, as the Gospel for the feast of Corpus Christi anticipates the canon of the mass. (Note that the roles of these elements in the baptismal pericope do not determine their relations in the smaller section — Mk. 10.20-34 plus the longer text, discussed above in pp. 1 65^6. Perhaps that pattern was worked into the pericope as a basis for yet more esoteric teaching.) (g) At this point Clement tells Theodore, to δε γυμνός γυμνω καί τάλλα περί ών εγραψας ούχ εύρίσκεται; in other words, the longer text as used in Clement's church

(or, as Clement chose to describe it) did not contain some material — including the phrase γυμνός γυμνω — which Theodore had reported as standing hereabouts in the Carpocratian text.

Since the Carpocratians had a reputation for sexual license (see Appendix B) and this section of the longer text reported that a youth came to Jesus περιβεβλημενος σινδόνα επί γυμνοϋ and stayed with him all night, it is easy to suppose that the Carpocratians took the opportunity to insert in the text some material which would authorize the homosexual relationship Clement suggested by picking out γυμνός γυμνω. Similar developments might be thought to lie behind the celebration of baptism in Acta Thomae 27 as η κοινωνία του αρρενος (cf. 132, text P, both lacking in the Syriac), and sayings like Gospel of Thomas (Leipoldt) 108, "Jesus said: 'He who will drink of my mouth will become like me, and I shall be he, and the hidden things shall be revealed to him.' " Cf. the longer text, εδίδασκεν γαρ αυτόν το μυστήριον της βασιλείας του θεού.

However, Clement does not explicitly say that the additional material was sexually offensive, and he would hardly have missed the chance to say so if it had been.

Therefore the γυμνός γυμνω probably belonged to a fuller account of the ritual.

Hippolytus, Apostolic Tradition XXI. 11, after speaking of the catechumen and the presbyter who is to baptize him, goes on to specify, "And let them stand in the water naked." The Carpocratian text may have contained some similar provision, more explicitly phrased.

Another possibility is indicated by the fact that the Carpocratians interpreted baptism as a resurrection. This appears from Irenaeus (Harvey, II. 48. 2 = Stieren,

II.31.2): After arguing that the Carpocratians perform their miracles by magic, and cannot perform major cures, he goes on to say, tantum autem absunt ab eo ut

mortuum excitent , quemadmodum Dominus excitavit et apostoli per orationem ( et in fraternitate saepissime propter aliquid necessarium , ea quae est in quoque loco ecclesia universa postulante per jejunium et supplicationem multam , reversus est spiritus mortui , et donatus est homo orationibus

THE SECRET GOSPEL

sanctorum) ut ne quidem credant hoc in totum posse fieri; esse autem resurrectionem a mortuis agnitionem eius , quae ab eis dicitur , veritatis. On this Harvey aptly quotes Tertullian,

De resurrectione mortuorum XIX. 2 ff. resurrectionem . . . mortuorum manifeste adnuntiatam in imaginariam significationem distorquent , adserentes ipsam etiam mortem spiritaliter intellegendam.

Non enim hanc esse in vero , quae sit in medio , discidium carnis atque animae , sed ignorantiam dei , per quam homo mortuus deo non minus in errore iacuerit quam in sepulchro.

Itaque et resurrectionem eam vindicandam , qua quis adita veritate redanimatus et revivificatus deo ignorantiae morte discussa velut de sepulchro veteris hominis eruperit . . . exinde ergo resurrectionem fide consecutos cum domino esse , quem in baptismate induerint. Hoc denique ingenio etiam in conloquiis saepe nostros decipere consueverunt , quasi et ipsi resurrectionem carnis admittant: " Vae," inquiunt , " qui non in hac carne resurrexerit ," ne statim illos percutiant, si resurrectionem statim abnuerint. Tacite autem secundum conscientiam suam hoc sentiunt :

"Vae, qui non, dum in hac carne est, cognoverit arcana haeretica ." Hoc est enim apud illos resurrectio. Sed et plerique ab excessu animae resurrectionem vindicantes, de sepulchro exire de saeculo evadere interpretantur , quia et saeculum mortuorum sit habitaculum , id est ignorantium deum, vel etiam de ipso corpore, quia et corpus vice sepulchri conclusam animam in saecularis vitae morte detineat.

Given this interpretation of resurrection as initiation (and vice versa) it is easy to suppose that the Carpocratians added to the ritual of their baptismal initiation some sort of resurrection ritual like that suggested by the parallels adduced in the commentary on γυμνός γυμνω above (on III. 13). Elisha's raising of the widow's son, II Kgs. 4.34, would have provided an OT precedent; its importance in Judaism of the Roman period is shown by its place on the front wall of the Dura synagogue

(panel WCi, Kraeling, Dura pi. LXIII), and the comments by impressed visitors (Geiger, Texts nos. 49, 51, 55). At all events, the one thing certain about the Carpocratian text is also the one thing most important for our present purpose: it is fully compatible with the interpretation of the secret ceremony as a baptism.

(h) After the baptism comes the sermon, the postbaptismal instruction, which still stands in canonical Mk. 10.35-45. The "message" is as follows: The newly baptized should not feel themselves at a disadvantage vis a vis Christians of longer standing, for not even the original disciples were assured the highest places in the kingdom. All who enter must drink of Jesus' cup (communion — the commemoration of the passion — will follow) and be baptized (as the initiates just have been) with his baptism (his death and resurrection), but for the future he who would be greatest should follow the example of the Son of Man who made himself servant of all and gave his life for us. So practice humility, make yourselves useful in the church, and give what you can.

With this concern for church discipline and finance the baptismal pericope comes to its natural conclusion. In the dramatic situation, that is, as put into Jesus' mouth, the references to the baptism and the cup in verses 38-39 must be understood as prophecies of his own passion and death and therefore of those of the sons of Zebedee.18

18. The amusing exegeses contrived to avoid this conclusion (Bernard, Study, Delling, Baptisma·, etc.) need not concern us; nor need the question whether or not the prophecy was fulfilled in the case of John. The writer of Mk. 10.39 maY have yet expected its fulfillment or may have erroneously believed it to have been fulfilled.

THE SECRET GOSPEL

But for that church which told the story, baptism and communion had become ritual means of participation in Jesus' death and resurrection (Rom. 6.3; I Cor.

11.24-28); so the question to the sons of Zebedee is now directed to all Christians, and their answer is the initiates' affirmation of faith in the magical identification with the saviour, effected by the rites of the mystery. To suppose the dramatic reference to the passion excludes the homiletic reference or the sacraments (Werner,

Einfluss 1 3 7ff ) is ridiculous. Equally ridiculous was the gnostic interpretation of the verses as referring to a second baptism (Irenaeus: Harvey, I.14.1 = Stieren, I.21.2), but it may have been accepted also by Clement's church in Alexandria (point a, above, p. 168).

This concludes the argument concerning the structure of Mk. 10.13-45 and its relation to the longer text. Since the argument has been complicated, it should be summarized here:

(a) Clement declares that the longer text was read in τα μεγάλα μυστήρια, probably the paschal service which included baptism.

(b) The liturgical use of the text would explain its importance to Clement, to the western text, to Lk., and to Mk.

(c) That τα μεγάλα μυστήρια were or included some sort of baptism is suggested by the baptismal reminiscences with which Clement introduces the text.

(d) The text stands at that place in canonical Mk. where baptismal material is to be expected.

(e) The details of Mk. 10.- 13-34 -f the longer text + 10.35-45 show the whole pericope to be a baptismal lection, to which all its parts are essential :

(i) 10.13-16 states the general prerequisite of baptism: become as little children.

(ii-vi) 1 0.17-3 1 states specific requirements: monotheism, obedience to the commandments, renunciation of property; these correspond to the standard pre liminary instruction for baptism.

(vii) 10.32-34, the credal prophecy of the passion and resurrection, provides both the assurance and the explanation of the rite's efficacy.

(viii) The " Gospel" for the service, the story of the raising of Lazarus, was from the second century on associated with baptism — both as a symbol of the initiates' resurrection from sin and as an example of the expected resurrection from death.

(ix) That the nocturnal initiation which follows the resurrection was understood to have been a baptism is argued by Ephraim's tradition concerning Lazarus and by the details reported: the six-day interval, the nocturnal character, the sheet and the nudity, and the communication of "the mystery of the kingdom of God," which seems to have been baptism.

(f) This baptismal interpretation explains both the inclusion of this material at this place in the longer text, and its omission from the shorter, exoteric text. It was included because the church for which the longer text was written performed baptism, as it performed the eucharist, by telling a story of what Jesus did. The story of the nocturnal initiation was the "canon" of the baptismal service, as the story

of the nocturnal supper was — and is— the canon of the mass. Hence the omission from the exoteric text, comparable to the omission of the "words of institution" from John. Hence also the doubling of the account of baptism — first symbolic (resurrec-

THE SECRET GOSPEL

tion) as the "Gospel" for the service, then actual (initiation) as the words accom panying the δρώμενα.

(g) The additions reported from the Carpocratian text also admit of baptismal interpretation, and this interpretation is made likely by reports of the Carpocratians' allegorizing of resurrection.

(h) Mk. 10.35-45 is best understood as postbaptismal instruction.

These arguments are cumulative. If even one of them be thought conclusive, it establishes the baptismal character of the section to which it applies, and thus strengthens the case for a baptismal interpretation of the adjacent sections. And even if none be thought conclusive, the fact that so many sections in sequence lend themselves so readily to the same interpretation will be difficult to explain away.

4. EVIDENCE FOR ABBREVIATION AT MK. 10. 46

For the second quotation from the longer text I see no such structural argument.

But it reports an incident located in Jericho, on Jesus' last trip to Jerusalem. Lk.

19.1-10 also reports an incident — the Zacchaeus story — and all the synoptics tell the story of blind Bartimaeus (doubled by Mt.) as something which happened near

Jericho. So there were stories connected with the place, and Bultmann is inclined to consider these local connections as "old tradition" or even "hochst primitiv"

( Geschichte 68f, 258, 364).

Moreover, the Zacchaeus story, like the preceding story in the longer text, is of a rich man who wished to see (cf. "be with") Jesus and therefore received him in his house and was saved. Other connections with the baptismal pericope in Mk. are his gift of (half of) his possessions to the poor (cf. Mk. 10.21) and the concluding sentence, ηλθεν yap 6 υιός του ανθρώπου ζητησαι καί σώσαι το απολωλός ; cf. the Markan conclusion (10.45, which Lk. omitted), καί γάρ ό υιός του ανθρώπου ουκ ηλθεν διακονηθηναι άλλα διακονησαι και δούναι την φυχην αυτοϋ λυτρον αντί πολλών. One is tempted to suppose that in telling the Zacchaeus story Luke had in mind the preceding

Markan pericope, including the passages from the longer text. This jibes with the evidence seen above that Matthew also knew the longer text (see the commentary on προσεκυνησε , II. 24; και προσελθών άπεκυλισε , III. 1 1 ; and the argument above on page 172), and the first of these bits of Matthaean evidence implies knowledge of the second addition, that following Mk. 10.46a.

Therefore, it is interesting that there seems to be a gap in the text of Mk. 10.46, just at the place where Clement locates his quotation from the longer text. 10.46a reads και έρχονται εις Ιερειχω. 10.46b reads και εκπορευόμενου αυτοϋ από Ιερειχω (and the text goes on to tell the Bartimaeus story). Notice the change of number between έρχονται and εκπορευόμενου αυτοϋ, and the repetition εις Ιερειχω . . . από Ιερειχω, which would be better understandable if something had intervened. Apparently the story of what happened in Jericho has been omitted. This appearance is strength ened by the way Matthew and Luke handled the Markan text: Matthew omitted

46a entirely, thus suppressing all reference to anything happening in Jericho; Luke changed 46a to εγενετρ δε εν τω εγγίζειν αυτόν εις Ιερειχω, omitted 46b, and told

THE SECRET GOSPEL

the Bartimaeus story as something which happened before Jesus reached the city.

He then went on, καί είσελθων δίήρχετο την Ιερειχω καί ιδού, etc. (19. if) and proceeded to tell his version of what happened in Jericho — the Zacchaeus story which, as we have just seen, has reminiscences of the baptismal pericope, including the longer text. So it seems that what happened in Jericho was something to which Matthew chose not to refer, and for which Luke had another, remotely cognate tradition which he preferred to Mk.

This appearance is strengthened by examination of the other passages where Mark begins a pericope with καί έρχεται/ έρχονται eis-. As noted above (commentary on II.23), this is one of Mark's favorite formulas and characteristic of his style as opposed to the other synoptists'. He used it in 3.20; 5.38; 6.1 ; 8.22; 10.1 (expanded) ;

10.46; 1 1. 15; 11.27; !4·32> and he used the same construction with other verbs or tenses in 1.2 1 ; 3.1 ; 5.1 ; 9.33. In all of these the formula is introductory, and in all except 3.20 and 10.46 it is followed by an account of some event which occurred in the place entered. The text in 3.20 was omitted by Matthew and Luke and looks as if it were the introduction to another deleted section of the longer text. Perhaps

9.33 another example of the same thing: καί ήλθον els Καφαρναονμ, , καί ev τη οΙκία γινόμενος. What happened to "them" in Capernaum before "he" went to "the"

(unexplained) house? According to Matthew, what happened was an encounter between Peter and a tax collector (the coin in the fish's mouth). Is this the only story in the Gospels to imply that Christians need not obey the civil law ? If so, the fact that it was kept secret is understandable.

Mark is known for the loose ends — references to unexplained houses or boats or people — his text contains (e.g., 3.20; 4.1; 6.32; 15.21) and for passages which look like abbreviations or references to omitted material, e.g., 1. 12-13; 3.6; 4.33f;

6.30,34; 7.31 (see Richardson's comment above, p. 161 n7) ; 10.1; 11.19; 16. ηϊ).

It is obvious that the text of canonical Mk. was abbreviated considerably by later revisers (Matthew and Luke), and other examples of abbreviation of Gospel material can readily be found (see above, page 94). Therefore it is not surprising that scholars of quite diverse views have supposed canonical Mk. produced by abbrevia tion: so Hilgenfeld, Markus ; Parker, Gospel ; Vaganay, Probleme. But such theories are necessarily speculative. In the present instance we have a specific case where expansion is reported by early tradition, but where the present text looks as if it has been produced by abbreviation and, as remarked above, traces of the abbreviated material seem to appear in Mt. and Lk.

The case for abbreviation is further strengthened by the fact that the omitted material mentions Salome. Salome appears in the NT only in Mk. 15.40 and 16.1

— in both as a witness of an event of great importance to Christian claims (the crucifixion, the discovery of the empty tomb; see the commentary on III.i 6). For these events she seems to have been one of the chief original witnesses, all of whom were women (so Mk. ; contrast the later Gospels).19 Yet Luke (23.55; 24.1) omitted

19. The way these women witnesses are introduced in Mk. I5-41 suggests that they were not previous ly mentioned in the Gospel as the author of 15.41 left it. But the introduction also shows that the author of 15.41 knew other stories about them. Were they stories he had cut out of the Gospel?

THE SECRET GOSPEL

both Markan lists of these women witnesses. [H.J.C. would prefer "rearranged," rather than "omitted"; cf. Lk. 8.3; 23.49,55; 24.10. But, though Luke did mention the other women elsewhere, he eliminated Salome's name.] Matthew deleted the name of Salome from the first list (27.56) and removed her figure entirely from the second (27.61; 28.1) and John also omitted her name (19.25). The Johannine text may come from a different tradition, but the omissions by Matthew and Luke were presumably deliberate. The presumption is that Salome was eliminated because persons of whom the canonical evangelists disapproved were appealing to her as an authority. For the late first century this is only a presumption. [H.J.C. compares

Lk.'s variants in the lists of the twelve.] But for the second century there is no doubt that Salome was popular in heretical circles; we have Celsus' report that the Carpocratians appealed to her authority (Origen, Contra Celsum V.62). Salome appears as one of the interlocutors of Jesus in the Gospel according to the Egyptians

(Hennecke-Schneemelcher, iogf), used by the encratites, Julius Cassianus, Theodotus the Valentinian, the Naassenes, and the Sabellians, as well as by II Clement and Clement of Alexandria. She also appears in the Gospel of Thomas (Leipoldt) 61b where she asks Jesus the curious question, "Who <(are you,) man, as from the one?

<(sic) You get onto my bed." (The text has, understandably, been corrupted.) Again in the Chenoboskion gnostic documents she is found in the First Apocalypse of James (Bohlig-Labib, KG A , p. 50). The Book of the Resurrection of Christ (James,

ANT 183) lists among the women who went to Jesus' tomb "Mary Magdalene, Mary the mother of James whom Jesus delivered out of the hand of Satan, Salome who tempted him, Mary who ministered to him and Martha her sister." Salome is one of the interlocutors in the gnostic Pistis Sophia (54, 58, 132, 145) and explains the mysteries. By contrast, the orthodox Ethiopic Didascalia (III.6) makes the apostles report that "there abode with us Mary Magdalene, and the sisters of Lazarus,

Mary and Martha, and Salome, and others also with them; (and) since He -(Jesus) commanded not them <sic, Harden) to teach along with us, neither is it right for other women to teach." (This occurs again in a Greek fragment of the Didascalia

Apostolorum III. 6, Connolly, p. xxvi, thinks Salome intrusive.) Of special interest is the double tradition found in the stories of the birth of Jesus. In the orthodox

Protevangelium Iacobi XIX.3-XX.4 Salome was not the Virgin's midwife, but a woman who heard of the virgin birth only after it had taken place and would not believe it. She attempted to test Mary's virginity and was punished by the withering of her hand. In the less respectable Liber de Infantia there were two midwives, Zelomi and Salome (as Amann notes, Protevangile 325, these are two forms of the same name).

Zelomi first made a manual test, proclaimed Mary's virginity, and gave glory to God; Salome refused to believe Zelomi's report, made a second manual test, and had her hand blasted. Finally, in a Coptic fragment quoted by Robinson (

Gospels ig6f) and in the Coptic Discourse by Demetrius on the Birth of our Lord (Budge, Texts

673ff), Salome was the only midwife; she immediately believed and became the first to proclaim the Gospel; and she followed the holy family and, later, Jesus everywhere, and— as Demetrius specifies— she saw everything. (A trace of this legend appears also in the History of Joseph the Carpenter VIII. 3: Salome accompanied

THE SECRET GOSPEL

the holy family to Egypt.) Obviously, Salome was a controversial figure. And it can be seen that the orthodox material has been edited to diminish her importance as a witness, for the oldest text of the Protevangelium Iacobi (which denies that she was the midwife) reports her cure with the words καί Ιάθη ή μαΐα lv rrj ωρα iKelvrj

(Testuz, ρ. ιο8, lines 9-Ι0> ca. a.d. 200). This shows reworking of an older form in which she was the midwife (a fact overlooked by Strycker, to say nothing of Testuz).

The hostile orthodox tradition is further represented by Origen on Mt., Commen tariorum Series 141 and 144, which admits that Salome watched the crucifixion from a distance, but denies that she was in at the resurrection — only Mary Magdalene and Mary the mother of James were admitted to that, "quasi maiores in caritate."

Against this, the semignostic tradition favorable to Salome is represented by numerous

Coptic texts on the assumption of the Virgin (Robinson, Gospels 51, 59, 60, 77) in which she appears as one of the Virgin's companions. (Some of these contain details reminiscent of the raising of Lazarus in the longer text.) Finally, in the Psalms of Thomas (note his connections) in the Manichaean Psalm Book II.222f, Salome appears as the equivalent of the OT "Wisdom" who builds her house, the Church.

[K.S. pointed out to me the similarity of Salome's disbelief and manual test of the virgin birth and Thomas' disbelief and manual test of the resurrection in Jn.

20.24-29. He also remarked (above, p. 166 ni4) that Jn. 20.24-29 plus Thomas' reception of the secret revelation (in the Gospel of Thomas) make up a rejectionresurrection-revelation complex similar to those in Mk. 8.29ff; io.2off; i4.27ff discussed above, pp. i65ff.] Of this complex John chose to retain only the first element and to use that as polemic against the followers of Thomas. Thomas had indeed seen, but was less blessed that the evangelist who, presumably, had not seen. (The explicit counterclaim in I Jn. 1.1 is less convincing than the implicit admission in Jn. 20.29.) Here, as in the development of the Salome stories, the original figures of Jesus' circle are pushed aside by the authorities of the developing churches. The importance of this will appear in the following chapter.

The above survey covers almost all the early Christian reports concerning Salome.

Omitted are only a set of notices in which Salome's name appears in lists of women who served the Lord or were related to the holy family or were involved in the resurrection stories. Such comparatively colorless material is directly derivable from the Markan references. Not so the tradition surveyed above. That tradition is almost entirely Egyptian (Strycker, 423, thinks Egypt the source of the Protevangelium

Iacobi) and is sharply divided between (a) orthodox polemic and (b) glorification by " heretical " material and by Coptic material which, although sometimes nominally orthodox, perpetuates quite unorthodox elements of Egyptian background. Since the Carpocratians who appealed to Salome's authority (Origen, Contra Celsum V.62) also maintained that Jesus was a natural man, the son of Joseph (Irenaeus: Harvey,

E20 = Stieren, 1. 25), and since Salome in orthodox material was cursed for her denial of the virgin birth, it would seem that she had figured as an authority for esoteric traditions allied with a naturalistic account of Jesus' birth, and that the importance of the esoteric traditions for the Egyptian churches had been sufficient to save her from the polemic which the naturalism engendered and to transform

THE SECRET GOSPEL

her into the first disinterested witness of the virgin birth. But whatever may be thought of the later history of the tradition, it is quite clear that the early material is so widespread and rich in content that it cannot be explained as derived from the two references in canonical Mk. There must have been other early traditions about Salome to explain the later developments. The later developments, in turn, suggest reasons for the suppression of the early material. That suppression is already visible in Mt. and Lk., and may therefore have operated in the editing of Mk.

E. Conclusions Let us now review the points which have been made.

A. The resurrection story in the longer text is an example of the miracle-inresponse-to-intercession story: its similarities in content to canonical examples of that type (e.g., the story of the Gerasene demoniac) are no more numerous than the similarities of the canonical examples to each other.

B. However, it is especially related to the story of Lazarus: the story in Jn. is a Johannine expansion of a later version of the story in the longer text. That the author of the longer text did not know the Johannine story is as nearly certain as anything based on source-analysis can be. That John did not know the longer text is probable.

G. When the longer text is added to Mk., the geographical outlines of Mk. 6.3216.8 and Jn. 6. 1-20. 2a are so similar and the order of major events and the places of these events in relation to the geographical outlines are so similar as to indicate that Mark and John independently expanded and reworked what were probably independent translations of some ultimately common source. Clement locates his first quotation from the longer text exactly where it should be in Mk. had it been part of this source. Also, the piece of geographical framework with which it concludes is paralleled in Jn., but differs so far from the Johannine parallel that it cannot be explained as derived from Jn. Presumably both it and its Johannine parallel came from the source. That the source was Aramaic is not unlikely in view of the numerous

Aramaic traits in the style of both Mk. and Jn., which have led a number of scholars to suppose them dependent, more or less directly, on Aramaic material (Burney,

Origin ; Black, Aramaic ; Schlatter, Sprache ; Torrey, Four Gospels and Translated Gospels , etc.).

D. The new material has particularly close structural ties with canonical Mk. : i. It fits the "historical" outline traditionally found in Mk.

2..Mk. 10.20-34 plus the longer text shows an order of events paralleled in Mk. 8.29-9.8 and Mk. 14.27-16.5. These parallels do not seem to come from a common source, but from common editorial arrangement of diverse material.

It may be that the same editor was responsible for all three instances of the pattern.

3. Mk. 10.13-34 + the longer text + 10.35-45 is a pericope of which the contents follow closely the order and contents of a baptismal service. Another

THE SECRET GOSPEL

form of the first section, related to the longer text, stood in a Gospel according to the Hebrews (Klijn, Question). The Markan (+ longer text) pericope was designed as a basis for baptismal teaching and for reading in the service. In this design the resurrection and initiation stories of the longer text are essential elements and occur exactly where they should. Moreover, the interpretation of the initiation as a baptism is confirmed by and explains the details given, and these explain Mk.

I4-51· 4. The second quotation from the longer text has been so bowdlerized by Clement or his predecessors that few conclusions can be drawn from it. However,

Mt. seems to have known it where Clement located it in Mk. Luke also locates in Jericho a story which has important similarities to the longer text (it looks like a late version of the same tradition with moralizing substituted for miracles).

The present text of Mk. seems to have suffered a deletion precisely where Clement locates the additional material. And the reference to Salome suggests a reason for the deletion.

Of the above points, B, C, and D concur to indicate that the longer text was the original text of Mk., and that canonical Mk. has been produced by abbreviation.

Against this conclusion must be set the conclusion reached from the stylistic study in the previous section (pp. 144145, above).

The evident conflict of these bodies of evidence leads one to examine what can be said against either side.

Against the stylistic evidence it can be objected that the excess of parallelism might be accounted for by the convergence of many different elements — the similarity in content of resurrection stories as such; the self-repetition of Mark; the use of liturgical and narrative formulas; the influence exerted by an important text; deliberate interpolation and manuscript corruption; and mere chance. And most important of all is the fact that the sample is too short to afford conclusive stylistic evidence.

Against the evidence from content it can be said that interpolation is often con cerned to produce texts useful for teaching, and interpolators often take advantage of small irregularities in texts interpolated. So the lacuna in Mk. 10.46 might have attracted an interpolator, and the concern to use earlier sections of the chapter for baptismal teaching, or for teaching of the vicarious atonement, might have led to their being filled out with appropriate material. Similarly, the fact that the resurrec tion story of the longer text shows a form older than the source of Jn.'s Lazarus story does not prove that the longer text was originally part of Mk. — early material could have been written down at a later date and interpolated in the canonical text. Thus the strongest piece of evidence for supposing the canonical text an abbre viation is the indication that the resurrection story stood in the common source of Mark and John. This argument could be eliminated by supposing John used the l93 longer text of Mk. ; but that supposition would have to be defended against the extremely strong case for John's independence of Mk., presented by Dodd in Historical

Tradition.

THE SECRET GOSPEL

If we suppose that John did not use Mk., but that both used different translations of a common source, it would still be possible to suppose that this common source was not used all at once. If the source was an esoteric document and if the first form of Mk. was written for beginners in Christianity — as Clement says — then the first form of Mk. might have been filled out at a later date by further selections from the esoteric source, and the editor who added these might have tried — particularly if he were translating from the Aramaic — to imitate as closely as possible the wording of the earlier Markan text. An Aramaic origin may be indicated by the attribution of another floating scrap of this material to a Gospel according to the Hebrews (Klijn,

Qiiestion) .

This theory seems to fit the evidence from style as well as that from content, but it still leaves unexplained the evidence that something was deleted from Mk.

10.46. However, there is no need to suppose that the editorial work on Mk. was limited to interpolation, or that all selections of the longer text — or all sections of canonical Mk. — had the same origin and history. The editor may have deleted as well as added; insertions may have been made by one editor and deletions by a second; other possibilities can easily be imagined.

Accordingly, if we were to shape our theory as closely as possible to the evidence, we should suppose that the latter part of Jn. and Mk. had as their remote source an Aramaic document which they knew in different translations and perhaps differ ently mutilated or interpolated or both. The earliest form of Mk., though using this source, did not include the resurrection and initiation stories now known from the longer text — although at least the former was in the source, and probably both were. However, a later editor cast these stories in a style constantly reminiscent of the Markan material he already knew, and added them to the Markan text. In doing this he was adding in written form material which had hitherto been kept secret and supplied orally in the teaching concerning baptism. The expanded text he produced was probably that used by Matthew. Whether it was later cut down again to form the present canonical text, or whether the canonical text was an older form which preserved its integrity alongside the interpolated text, there is no telling.

But in 10.46, at least, the present canonical text seems to be an abridgment; and the longer text, as quoted by Clement, preserves the introductory phrases of the material cut out.

Postscript: Rereading this text in 1970, more than seven years after it was written and four years after its revision, I find the argument from style much weaker than that from content. Also I notice that I have not considered the likelihood that Clement, who had no reason to love the secret Gospel, might have been inclined to prefer an account representing it as a secondary expansion of the shorter text which in his day was well on its way to becoming "canonical." Perhaps, therefore,

I have overestimated the reliability of his report. But the theory proposed above still seems to me the one which would best fit the evidence reviewed.

FOUR Background A draft of this chapter was read by Professors Cyril Richardson and Gershom Scholem, to both of whom I am indebted for helpful discussion.

III. Reports about the secret teaching of Jesus, 199 IV. Questions about the content of Jesus' secret teaching, 201

VI. The problem of Jesus' role in relation to the present kingdom, 204

he kingdom of God,

A. The rite was a means of uniting with Jesus, 2 1 7 B. The union was effected by the spirit, 219

C. The rite was magical, 220

Secrecy in ancient Judaism, 197 1. The term and the facts, 220 2. The question of spells, 222 3. Minor magical traits of the miracle stories, 223

4. The predominantly magical character of the Gospel stories, 224 5. The relation of "magician" to θείος άνηρ and "Son of God," 227

6. Jewish and pagan opinions of Jesus, 229 .

The question as to the historical value of the new Gospel text, 196 7. Magic in the practice of Jesus' followers, 231

8. Recapitulation and conclusions, 235 D. The rite was a means of ascent to the heavens, 237

the Baptist,

1. Background, 238

ding to Paul, 213 2. Indirect evidence of Jesus' practice, 240 3. Records of Jesus' practice, 243 a. The transfiguration, 243 b. Phil. 2.5-1 1 ; I Tim. 3.16; Jn. 3, 244

E. The rite liberated its recipients from the Mosaic Law, 248

A. The coming of the spirit, 252 B. The adoption and development of baptism by the early churches, 253

C. The libertine tradition in early Christianity, 254 I.

Evidence from the Gospels that Jesus baptized, 209

Elements derived from Jesus in Pauline baptism, 216

CLEMENT OF ALEXANDRIA

1 . The persecution of the early churches, 254 2. The parties in the early churches, 256

3. Direct evidence of libertinism, 258 4. Source and dispersion of the libertine tradition, 262

D. The "loss" of all writings of Jesus'1 immediate followers , 263 E. Second-century silence about Paul's doctrine of baptism , 264

XII. Conclusion: The historical value of the new text, 265 XIII. Carpocrates, 266 A. Names, 266 B. Date, 267 C. Testimonia, 268 D. Studies, 276 I. The question as to the historical value of the new Gospel text Solution of the literary problem as to the relation between the new Gospel text and canonical Mk. does not solve the historical problem as to the reliability of the material contained in the new text. Since all first-century Christian authors drew on the traditions of their churches, as well as on their own imaginations, an early text may contain mere invention or a late one report old tradition. Therefore, although some of the material in the longer text probably came from a Gospel earlier than canonical Mk., and although some of it was probably cast in its present form a decade or so later than most of the stories in the canonical Gospel, neither of these probabilities is of much value as evidence of historical reliability. False statements may be primitive, and true statements may be secondary accretions to original errors.

Given this state of affairs, and given the evidence collected by Bultmann, Geschichte, passim, it would be naive to ask whether or not the events reported in the longer text "really happened." In dealing with the Gospels we have no prima facie criterion of truth save verisimilitude, and verisimilitude is not reliable. But we can ask how the new text is related to the historical problems which have been raised by the study of the canonical Gospels.

One of the most important of these problems is that of the source and significance of the secrecy motifs in the Gospels. With this we may conveniently begin, especially because Clement's letter, which quotes the new text, has revealed that Clement's church in Alexandria kept this Gospel secret and read it only to those being initiated into the "great mysteries." Moreover, the new text itself reports that Jesus adminis tered to one of his followers a nocturnal initiation in which "Jesus taught him the mystery of the kingdom of God." How is this report related to the canonical Gospels' reports of Jesus' secret teaching?

THE BACKGROUND

II. Secrecy in ancient Judaism First, the background: Throughout the ancient world secrecy was practiced everywhere — in government and politics, in the trades and professions, in the philosophic schools, and, above all, in religion and magic. Secret rituals had been particularly common in ancient near eastern religions (Hooke, Religion 53), and the religion of Israel had been no exception in this respect: its official center was a temple of which the area around the altar might be entered only by members of the various grades of the priestly caste. None save the high priest himself was per mitted to enter the adyton (Lev. 16; Num. 1.47-54; 4.1-15; 8.14-19; 17. 1-5;

18.1-7; etc.).

Judaism not only perpetuated but developed these esoteric traits. The main court of the temple was now closed to gentiles (Bickerman, Warning 390-394, remarks on the similarity to pagan mysteries), and increased emphasis on purity law did much to cut off Jews from their gentile neighbors. The degree of separation has been greatly exaggerated in such works as Moore's Judaism , but the fact of it is undeniable. Consequent, the gentiles saw the religion of the Jews as a mystery religion and the Jews themselves represented it as such (see above, pp. i8off).

Besides the assimilation of Judaism in general to the mystery religions, the several sects which grew up within Judaism kept their doctrines and rites secret from each other and from ordinary Jews. These sects probably originated in differences of legal theory and practice, particularly differences concerning the purity laws (see my Sect). Consequently, their meals, their houses, their schools, and their synagogues were apt to be closed to all "outsiders" ( τοΐς e£o>, Mk. 4.1 1), gentile and Jewish alike. (Accordingly, Strack-Billerbeck on Mk. 4. 1 1 is mistaken. The commentary there contains only one really relevant passage — Megillah IV. 8— and that one it misinterprets. The DTiirn are not "ketzerisch gerichtete Menschen"; on the contrary, they are sharply contrasted with the heretics, who are the adherents of mrft; the cm^rn are the ordinary Jews who are not members of the Pharisaic sect or of any other , while the heretics are members of competing sects — see the notes of Bertinoro and Yom Tov. But the precision here is determined by the contrast.

When there is no such contrast, DTi^T! means nothing more or less than " those outside"; what they are outside of must be determined from the context.)

Groups cut off from the outside world by such legal barriers usually developed further peculiarities of doctrine and observance, and among these was apt to be a deliberate affectation of secrecy concerning their teachings and practices. How far this was rooted in childish delight at having secrets, how far it was based on practical considerations of discipline and prudence, how far it was influenced by the examples of philosophic schools and mystery cults are questions which doubtless had different answers for each sect. Anyhow, it is clear from the disciplinary material in the Qumran finds that the pattern was well established in Palestine before the time ol

*97 Jesus. See, for instance, Manual of Discipline IX.i6f: It is a legal obligation to conceal the sense of the Law from wicked men. Again, VI. 13— 23: An initial examination and covenant is required before the outsider can even be instructed in the rules ol the

CLEMENT OF ALEXANDRIA

sect, two years' probation before full admission. This material roughly agrees with the descriptions of the Essenes' practice of secrecy as found in Josephus, BJ II. i igff and Hippolytus, Philosophumena IX end (on which my Description) , also Josephus,

AJ XVIII. i iff. (That Jewish laws were to be kept secret, a fortiori, from the gentiles, is a notion probably older than the Essenes and more widely accepted than their practices; it appears already in the Testaments of the Twelve Patriarchs , Judah XVI. 4.)

These agreements help secure credence for the account of the secret practices of a similar group in Egypt given by Philo in On the Contemplative Life. A society of would-be revolutionists whom Josephus describes as another "philosophical" sect

(. BJ II. 1 18; AJ XVIII. 23; etc.) must certainly have practiced a good deal of secrecy for prudential reasons; so must later revolutionary groups, of which the most famous were to be the "Zealots" (with whom Jesus' zealous disciple Simon

(Lk. 6.15; Acts 1. 1 3) had probably no connection). The tradition of doctrine and practice from which came the Merkabah literature (especially the Hekalot tracts) and the "Mithras Liturgy" (discussed above, p. 181) was certainly secret and must have existed in Palestine during Jesus' lifetime. The priests, particularly the upper priesthood, must have had a large body of secret traditions and practices; so must the Samaritans (among whom, in turn, were a number of secret sects). We hear of the Samaritans keeping secret, for instance, the principles by which they regulated their calendar (as did the Dead Sea Sect: Bowman, Calendar 27ff). But of all these groups we are badly informed. The one sect best known, the Pharisees, practiced secrecy in many fields. Jeremias ( Jerusalem II.B.io6ff) has shown they had secret doctrines about God, his name and his throne, the heavens, creation, the structure of the world, eschatology, the "reasons for the Law," sexual questions, magical formulas, discreditable traditions, and laws or legal provisions likely to be abused.

He remarks that this "Arkandiziplin" played an even larger role in early Christianity, where it also involved Ghristology, the secrets of the divine nature, and the sacraments

(ibid., 109).

The material Jeremias collected could easily be increased (see, e.g., Jervell, Imago 2 of, 72f, etc.). Moreover, these secrets are secrets which were kept even from ordinary members within the Pharisaic sect. Therefore, besides all this, it must be emphasized that vis-a-vis Judaism at large the Pharisaic sect itself was, in the first century a.d., an esoteric group. Melamed, Leshe'elat 118, shows that public teaching was prohibited (except as a demonstration in times of persecution) by the older interpretation of the Law (which Rabbi tried to maintain as late as the end of the second century). This secrecy produced in rabbinic material a form of " pronounce ment story" which is also important in Mk. : An outsider asks a question of a rabbi, the rabbi gives him an inadequate or even false answer, later the rabbi's students ask him in private about the matter, and he gives them the true explanation; e.g.,

Bereshit Rabbah 8.9 and parallels. See, for example, the stories about Yohanan ben

Zakkai, a younger contemporary of Paul and Jesus, in Bemidbar Rabbah 19.4; Tanhuma '

(ed. Buber) Hukat 26; J. Sanhedrin 1.2 (19b). (In the first group of these passages his secret teaching is that a dead body does not render the man who touches it impure, nor does the immersion legally prescribed for purification after contact with dead

THE BACKGROUND

bodies purify; but the laws are to be observed as arbitrary divine rulings. Cf. Jesus' saying, Mk. y. 15, ουδόν όστιν e£co6ev του ανθρώπου είσπορευόμενον εις αυτόν ο δυναται κοινώσαι αυτόν. This in Mk. is followed by a private explanation to the disciples.

On the story of Yohanan see Daube, JVTRJ 1 4 1 fT; Neusner, Life 6 if.) III. Reports about the secret teaching of Jesus

Stories of this same form — pronouncement, question in private, secret teaching to disciples — are told of Jesus in Mk. 4. ioff; 7. off; 9-28f; io.ioff; 13.3 fT. The private questions and answers in these passages have often been attributed to the redactor of Mk. : e.g., Bultmann, Geschichte 356; Seitz, Criteria. The latter (p. 220) affords an amusing example of "historical" criticism: it proves the secondary character of Mk. 7. 1 8ff by the fact (!) that a special revelation from heaven was required to persuade Peter to go with a gentile. Admitting on literary grounds that the Markan secret explanations are redactional, one has to ask why the redactor chose to represent

Jesus in this way. It is not enough to argue (as Seitz does) that the content of the secret explanations represents the wishes of the gentile church. Even if one were to accept the dubious distinction between the "primitive Palestinian" and "gentile" churches, one would have to explain why the secondary material was introduced as secret teaching. Why could the question not have been asked from the crowd and the answer given openly? To this the customary reply, since Reimarus, has been that doctrines falsely attributed to Jesus had to be presented as secret teaching because his public teaching was known and the intended dupes would remember that he had not publicly taught such things. But, first, it is unlikely that the public teaching of Jesus was so completely known to readers of the Gospels that they could confidently deny it had contained one or another element. And, second, if it be supposed that Jesus' teaching was clearly and fully remembered by the churches, they would have remembered also whether or not he taught in secret. So the fact that the editors of the Gospels chose to present some additions as secret teaching suggests their churches had a tradition that Jesus did teach in secret, and this tradition was older than the Gospels which relied on it to lend credit to their accounts of what he thus taught.

This leads to the question, are there other traits in Mk. or in other early Christian documents which confirm these reports of secret teaching?

It is well known that there are such traits in Mk. Wrede has left a classic description of those which, as he says ( Messiasgeheimnis iqbff), were noticed by many critics —

Schleiermacher, Strauss, Keim, Hilgenfeld, and others — before his time: the taking aside of the sick, the manipulation and use of tangible means in the miracles, the magical overtones, the conception of miracles as mysteries, the conception of the disciples as initiates to whom the mystery has been given, the conception of Jesus

!99 as a god disguised in flesh, his mysterious, secret journeys, his traveling incognito, his going apart alone or with a few disciples — "scenes from the life of a magician."

Moreover, Wrede declares, the earlier critics were right. All these are characteristics

CLEMENT OF ALEXANDRIA

of Mk. as opposed to the other Gospels. The earlier critics saw them because they looked at Mk. as it was; the later critics overlooked them because they came to Mk. with the preconceived notion that it was the earliest and most historical Gospel, the repository of Peter's memories, and therefore looked at it only to find material which would pass as historical and possibly Petrine. " Schleiermacher has already remarked that this Gospel approximates the characteristics of the apocryphal ones.

How he meant this, is not here in question. But one thing seems to me certain:

If this Gospel came to light today , for the first time , in some grave , Schleiermacher' s judgement would be approved by more than a few." {Ibid., my italics.) Cf. Schille, Formgeschichte 18:

"Why does Mark choose, in developing Jesus' speeches of revelation, a form which presumably was customary in the catechetic initiation of neophytes in the teaching of the Church, as if Jesus like a hierophant led the twelve and 'those around him' step by step into the mysteries of the i Church' ?" (His italics.) Even Wrede's enumeration does not exhaust the esoteric traits — there are, further, the teaching in parables, the dis ciples' failures to understand, the commands of secrecy, the actions performed before a few chosen witnesses, the contrast of the immediate circle with those outside, and so on. The early Christians recognized these traits and consistently pictured Jesus as teaching mysteries in secret, Cramer 31 1, 335, 353, etc.

Since Wrede's time this material in Mk. has commonly been discussed under the heading "the messianic secret." This is a pity because, as Wrede's list shows, the traits are too various to be explained by the sole secret that Jesus was the Messiah.

What has that to do with — for instance — the use of magical techniques in the miracles ?

A further misfortune resultant from Wrede's brilliant study has been the tendency to treat these phenomena as peculiar to Mk. (See the review of theories in Boobyer, Secrecy

225k) This was not Wrede's fault. He had examined the other Gospels, too, and had found in Mt. a different theory of a secret teaching, exemplified in such passages as g.27ff; 11.25 aRd 1 2. 1 8flf ( Messiasgeheimnis 151-162) and in Lk.-Acts a theory closer to Mk.'s (i65ff). Here he noted especially the report of Jesus' postresurrection teaching (Lk. 24.25^ Acts 1.3, τά 7 τερι της βασιλείας του θεοϋ) but did not see that this was intended to authenticate some body of secret doctrine known to Luke. He also observed (i88ff) traces of the Markan theory in Jn.

Jesus' claim in his trial, εν κρύπτω ελάλησα ονδεν (Jn. 18.20), probably reflects a charge of secret teaching. Neglecting this claim, Jn. represents Nicodemus as coming to Jesus for secret instruction by night (eh. 3 — the parallelism to the longer text of Mk. is clear), reports that Jesus had secret disciples (19.38^, and makes the last supper a long secret lecture. Also, the disciples in Jn., like those in Mk., repeatedly failed to understand: 2.22; 12.16; 13.7,28; 14.20; 16.12,25; 20.9. Only after Jesus' death came the spirit which was to recall to them all things Jesus had said (i4.i6f) and lead them into all truth (16.13); that is, into a revelation more revealing than

John's Gospel, of which the last speech was still εν παροιμίαις (16.25).1 This again i. This closely resembles the relation which Clement said existed between the longer text of Mk. and the secret teachings: the longer text contained those stories and sayings the spiritual inter pretation of which would lead the initiates to the hidden truth (Letter 1. 2 1-26, and commentary, also p. 166 above).

THE BACKGROUND

must be a claim to secret teaching or to private revelation which would immediately pass over into secret teaching (cf. Kragerud, Lieb lings j linger 84ff; Kasemann, Ketzer 302).

The identical claim is found in Paul, I Cor. 2.1 1, and there is no doubt that Paul has secret teachings which he will not reveal even to baptized Christians who are still "carnal" (ibid., 3. iff). Machen, Origin, has made a strong case for the supposition that the main elements of Paul's religion came from Jesus. We saw above (pp. iy8ff) that Paul and Mk. were at one in their use of μυστήριον to refer to baptism. They are also at one in their belief that the teachings of Christianity include secrets to be revealed only to the few elect. A further point of agreement is their belief that Jesus was a supernatural power in disguise: Mark thought demons could recognize him but men could not (1.24,34; 3.1 1 ; etc.) ; Paul thought even demons were fooled

(I Cor. 2.8, on which Macgregor, Principalities', cf. Phil. 2.7; Col. 2.15; Jn. 19.11);

Paul's opinion was shared by the first-century ( ?) author of the Ascension of Isaiah

10— ii. (A similar belief was probably held about Simon Magus by his followers:

Acts 8.10; Cerfaux, Gnose, 1. 504b See the remarks of Weiss, Christianity 758b) In sum, the Gospels all represent Jesus as teaching in secret; Paul certainly had secret doctrines and Lk. and Jn. presuppose them. When Christianity first appears in the writings of pagan authors it is described as a secret society (Pliny, Epistulae ad Traianum 96.7: hetaeria', Grant, Pliny, PW sub voc.) or an initiation (Lucian,

Peregrinus 1 1 : τελετή) . These descriptions are completely in accord with its background in sectarian Judaism. Accordingly, we need not question the report in the new text that Jesus taught in secret, but we may reasonably inquire what he secretly taught.

IV. Questions about the content of Jesus' secret teaching But can we hope, at this remove, to discover what Jesus taught secretly? There is no scholarly agreement even as to what he taught in public. He has been repre sented as a rabbi, a philosopher, a pacifist, a revolutionary, a moral reformer, a prophet of the coming end of the world, an itinerant exorcist, and the son of God, come down to earth to reveal his own nature. This general disagreement as to his teachings suggests that some of them, at least, were secret. So does the cause of the general disagreement — the fact that the canonical Gospels contain considerable elements which give apparently contradictory pictures of their hero. These contra dictions must be dealt with either by supposing one body of material "primary" and the rest "secondary" (the method which has produced the various "historical pictures of Jesus listed above), or by supposing them all "secondary" (an act of faith, not to say credulity), or by supposing that they reflect different facets of a complicated character in which they were reconciled (more or less) by considerations not made public. Of these three methods the first and last are not mutually exclusive, but their possible combinations do not here concern us. Here the facts to be noted

CLEMENT OF ALEXANDRIA

are that any attempt to explore the area of secrecy in the Gospel tradition is neces sarily speculative, but any attempt to deny it is necessarily wrong. Even texts like

Mt. 10.27, "What I tell you in darkness, declare in the light, and what you hear in the ear, proclaim on the housetops," while intended to suggest that (all?) the content of the secret teaching has now been made public, admit implicitly that a secret teaching existed. Accordingly we must recognize both that we deal here with an extremely speculative area of study, and also that the preserved evidence necessitates speculation.

To the question "what did Jesus tell his disciples in darkness?" the answer has usually been "the messianic secret." This secret is divided by a recent study (Burkill,

Revelation) into two parts: the secret fact — that Jesus was the Messiah — and the secret interpretation of the messiahship as a career of service, suffering, and death.

As for the secret fact — it seems likely that Jesus did think he was the Messiah, and had obvious, prudential reasons to conceal his opinion. As for the secret interpretation, however — there is no clear reason why that should be kept secret, and Mk. insists that Jesus taught it publicly (8.3 if). Moreover, it seems unlikely that he ever taught it at all: the prophecies of the passion look like prophecies ex eventu — the weakness of the opposite opinion has been demonstrated by Taylor's defense of it ( Origin ).

And the report that Jesus set guards at Gethsemane (Mk. 14.32,34; cf. Lk. 22.40) indicates that he had no intention of giving his life as a ransom for any. The question of Shemtov ben Shaprut, reported by Krauss (

Leben 269) is worth repeating: If Jesus gave himself freely to his sacrificial death, why did he say that Judas Iscariot betrayed him?

On the other hand, there is no reason to suppose Jesus had only one secret— the fact that he was the Messiah. The existence of other secrets may be indicated by obscurities in the tradition. One such obscurity is that covering the relation of the Messiah to the kingdom of God. And since Mk. 4.11 declares that "the mystery of the kingdom of God" has been given to the disciples (as opposed to "those outside"), and the new text represents Jesus as teaching this mystery to the youth who came to him for nocturnal initiation, we seem to have here another element of Jesus' secret teaching. We have already seen evidence indicating that the mystery was a baptism (above, pp. I78ff). We must now try to find out what this baptism was supposed to effect, why it was administered by Jesus, and why it was secret. But these questions presuppose another: If baptism was "the mystery of the kingdom of God," what was "the kingdom of God"?

V. The kingdom of God The discussion of the kingdom touched off by Dodd's Parables has practically ended with the recognition that "kingdom" meant primarily, "rule" (as rVDVft

— an abstract — would) but might, by extension, refer to the persons or organization or area ruled ; cf. Lundstrom, Kingdom and Perrin, Kingdom. Of recent articles, Ladd,

THE BACKGROUND

Reign, is right as to the essential meaning but does not allow sufficiently for the extensions. Grant, Idea 442, gives a better account by comparing God's rule of all creation to the rule of the Great King: it had originally been complete, but since the time of Adam certain provinces (the demonic, human, and animal worlds) had been in revolt, and in Jesus' time apocalyptic writers had recently been foretelling that the revolt was soon to be suppressed and the rule restored. It was, later, the peculiarity of Jesus' followers to believe that the suppression had already begun; the Great King's rule had come back into the revolted provinces in the person of his representative, Jesus, it had been manifested in Jesus' acts of power, and it was continued in their own obedience, as they looked forward eagerly to his coming again "with power and great glory" to complete the restoration. This explains why the NT documents generally speak of the "kingdom" both as present and as future — it was both at once: It was present in Jesus, in the Church, and in God's eternal rule of the heavens; it was yet to come in the full resubjugation of the lower world (so Kiimmel, Eschatologie) .

That the Church is a manifestation of the "kingdom of God" confirms our previous interpretation of "the mystery of the kingdom" as baptism — the ritual of initiation into the Church. Stanley ( Kingdom ) has shown that in Mt. and Lk. the kingdom is sometimes the Church (so especially in Mt. 13.33,52), and the same sense appears in Paul (Col. 1.13; 4.1 1 ) , the Apocalypse (1.6,9, etc.), and the Markan parable of the mustard seed (Mk. 4.30; cf. Dan. 2.35,44 and Jeremias, Gleichnisse 93).

A further point made clear by Grant's analogy is that God's rule of the heavens continued unchanged in spite of the rebellion of the lower provinces. God's throne either is in or is the heavens (Apoc. 4.2; Mt. 5.34; 23.22) and God himself is in the heavens (Mt. 6.1,9, etc.); the heavens are therefore his kingdom, κατ' εξοχήν. This had been the opinion of the Psalmist (103.19, where } is to be translated "even though" — cf. 115.16). In Wisdom, too, "the kingdom of God" is in the heavens, where it was shown to Jacob in his dream (10. 10) and whence the divine word descended to destroy the wicked (18.15). So, too, in the pseudepigraphic apocalypses,

God is customarily enthroned in the heavens and his kingdom is, by implication, there; the implication is made explicit in III Baruch (the Greek apocalypse) ii.if, where the keys "of the kingdom of heaven" are identified as those to the gate of the fifth heaven. Compare Dan. 4.34, where God is "the King of the heavens." This explains the fact noted by Aalen, that in the NT "the kingdom of God" often has a local sense and evidently refers to "a confined area" (

Reign 229). This "confined area" is in the heavens, the realm of God, as Riesenfeld noted (ΠΑΡΑ). There "in the kingdom of God" is Abraham (Mt. 8.1 1, Lk. 13.28), to whose bosom Lazarus was carried by angels immediately after death (Lk. 16.22; cf. Dodd, Parables 44).

There (in the third heaven) is the Paradise to which Paul was taken up on his conversion (II Cor. 12. 1-4; cf. Gal. 2.1), to which four second-century rabbis reportedly ascended (J. Hagigah II. 1 [7 7t>] )

, and to which Jesus promised to take,

on the very day of their deaths, the thief who believed that he would yet come into his kingdom (Lk. 23.42^. It is in the heavens that the reward of the righteous is laid up — the kingdom prepared for them from the beginning of creation (Lk. 12.32-

CLEMENT OF ALEXANDRIA

34; Mt. 5.12; 6.20; 25.34). It is in the heavens that God's kingdom has come, that is, his will is done, already, as the Church prays it may come on earth (Mt. 6.10).

So too the Synagogue prays, in a probably contemporary prayer: "May he who maintains peace in his heavens (ΤΊΟΙΊ^Π) maintain peace for us and for all

Israel" (Baer, Seder 104). The keys of the kingdom of the heavens, which are to be given to Peter, will render his legal decisions binding in the heavens; compare the rabbinic claim that God's court in the heavens confirms the actions of rabbinic courts on earth ( B . Makkot 23b and parallels, etc.). Before Abraham came, says

Sifre Devarim 313 (on Dt. 32.10), God was, as it were, King only over the heavens; when Abraham came God began to rule also on earth. By taking away the key of knowledge the experts on the law locked the kingdom of the heavens in men's faces; they neither went in themselves nor permitted others to enter (Mt. 23.14; Lk.

11.52; P. Oxy. 655; Kraft, Oxyrhynchus; Gospel of Thomas (Liepoldt 39); Clementine

Homilies XVIII. 15-16). And finally the author of II Tim. 4.18 prays that the Lord will deliver him from every evil thing καί σώσει els την βασιλείαν αύτοΰ την επονράνιον.

This demonstration that "the kingdom of God" may refer to a locality in the heavens has been protracted because the fact is usually ignored. Having established the fact — for which we shall later on bring further evidence — we can now return to the previous problem : what was Jesus' secret teaching about baptism, "the mystery of the kingdom"? Here we shall try to determine Jesus' position by comparison with the Baptist's on the one hand and Paul's on the other.

The interval between the work of the Baptist and the preserved letters of Paul is a mere 25 years (ca. a.d. 25-50). Therefore the work of Jesus can be defined func tionally as that which, beginning from the Baptist, led to Paul. Accordingly, after some preliminary remarks on the difficulty of distinguishing Jesus' role from the Baptist's, we shall define as sharply as possible the work of the Baptist, then review the NT evidence that Jesus baptized, and then consider Paul's statements about baptism and try to discover in them the elements which derive from Jesus.

VI. The problem of Jesus' role in relation to the

PRESENT KINGDOM

According to the Gospels, the important difference between Jesus and the Baptist was that Jesus was the Messiah, the Baptist, merely a forerunner of the Messiah.

The Gospels are clear as to the role of Jesus in the future, when the kingdom will come "with power": then he, as Messiah, will be the chief executive (Mt. 25.3 iff;

Mk. I3.26ff; etc.). But in the present kingdom, the kingdom on earth in his day, he is generally represented as an advance agent: his function is to proclaim the coming of the kingdom and to exemplify its presence, to manifest its power in his

miracles and its requirements in his preaching (so, for example, Dibelius-Kummel,

Jesus). But all this could have been done by any prophet and therefore, a fortiori,

THE BACKGROUND

by the Baptist, who was admittedly more than a prophet (Mt. n.gff; Lk. 7.26ff. —

Jn. 10.41 is presumably partly polemic). Therefore we must ask whether Jesus thought his present role different from the Baptist's, and, if so, how.

It will do no good to say that with the Baptist the kingdom was yet to come but in Jesus' work it was present. For we have seen above that the kingdom is simply the rule of God, therefore it is present whenever God manifests his power or men obey him. So if the presence of the kingdom means only its presence in preaching and prediction and acts of power, it was present already in the work of the Baptist.

And even if we were to suppose Jn. 10.41 correct in reporting that the Baptist did no miracles (an unlikely supposition), and if we should follow Mt. n.2ff || Lk.

7. 1 8ff and Lk. 11.20 in supposing that Jesus saw in his own miracles the proof that the kingdom was coming in a new way, we should have still to ask what consequences he drew from this belief. Did he think his present role was merely to preach, prophesy, and perform miracles? And did he expect no response but belief, repentance, and expectation of the great change ? (These were the responses expected by the Baptist,

Mt. 3.1-12 and parallels.) Or did Jesus think of himself as having some further function; did he think there was something which he and his hearers could do , but which the Baptist and the Baptist's hearers could not?

Like the Baptist, Jesus was to be executed by the civil authorities and to rise from the dead (Mk. 6.16; 8.28), but he probably did not foresee these details of the divine plan until the last minute; therefore the Gospels contain little material explaining how the crucifixion was useful for or relevant to the kingdom, and what little they do contain probably dates from the period after Jesus' death. Another peculiarity of Jesus' work was the ceremony of the last supper, but this was reportedly performed only once, on the last evening of his life, and therefore does not represent the sort of thing we are looking for — something which the Messiah (as distinct from a prophet) could do for his followers, some action which his followers could take because He

(and not merely another prophet) had come. It is not impossible that, as Schweitzer believed ( Leben-Jesu-Forschung 42 if), the stories of the feeding of the multitude reflect some sort of symbolic anticipation of the banquet of the righteous in the world to come; but the evidence for this interpretation is not convincing. So the problem as to Jesus' messianic function, in relation to the present kingdom, remains open.

VII. The role of the Baptist The problem of Jesus' function is the more acute because the Baptist certainly did have a notion of his own special function, by which he was to prepare his hearers for the coming of the kingdom, and of a special response (beyond belief, repentance, and reformation) by which his hearers could be prepared. His function was to

administer a new rite, a "baptism of repentance for the remission of sins" (Mk.

1.4 and parallels).

CLEMENT OF ALEXANDRIA

From the NT it can at once be seen that the Baptist's work marked an epoch for early Christianity. Not only do Mk., Q, and Jn. represent it as "the beginning of the Gospel" (Mk. i.iff and parallels; Lk. 16.16b; cf. Jn. 1.6), but it seems to mark in fact the beginning of the Christian tradition, which contains no reliable report of anything earlier. Moreover, it is repeatedly referred to, in the traditional material, as the point from which the movement began: "The law and the prophets were until John." "From then on" is the Christian period (Lk. 16.16 || Mt. ii.i2f).

When a successor is chosen for Judas he must be one who was a member of the group throughout all Jesus' career "beginning from the baptism of John" (Acts

1.22). When Peter begins to explain the Gospel to Cornelius and company he assumes,

"You know what has been happening throughout all the Jewish districts, beginning from Galilee, after the baptism which John proclaimed" (Acts 10. 37). 2 When Paul presents the Gospel to the Jews of Pisidian Antioch he dates Jesus' work "after John had preached . . . the baptism of repentance to all the people of Israel" (Acts 13.24).

Moreover, from what the Gospels tell it can be seen that the Baptist's importance was not limited to the Christians. All Judea and Jerusalem (Mk. 1.5) and Transjordan

(Q, Mt. 3.5 II Lk. 3.3) went out to him, in crowds (Mt. 1 1.7 || Lk. 7.24; Lk. 3.7,10; etc.). His teaching was accepted by masses of the common people (Lk. 7.29). Even the authorities of the Temple are said to have been afraid to speak against the Baptist, even after his death, because of his popular following (Mk. 1 1.32 and paral lels). His disciples carried his sect as far as Alexandria and Ephesus (Acts 18.25;

19.3). In this matter the testimony of the NT is confirmed by Josephus, who had heard of John as a figure both influential (Herod Antipas had him executed for fear he might initiate a revolt) and popular (many Jews interpreted a subsequent defeat of Herod's army as a divine punishment for this execution, AJ XVIII. 1 i6ff).

The importance of the Baptist makes it not unlikely that he is the object of the polemic in the Qumran Manual of Discipline III.4ff against the notion that baptism can remit sins. The opponent is not named, but the passage seems to have one and to insist, against him, that men can be cleansed of their iniquities only by the holy spirit and by submission to the rules of the sect (III. 7-8), and only after this spiritual cleansing can their flesh be cleansed of impurity by the regular OT method of sprink ling with water containing the ashes of a red heifer properly killed and burned

(Num. 19.9,13, etc.).3 That the object of the polemic was the Baptist is made very likely by the fact that Josephus, in the passage cited above, saw fit to defend him against such charges and to insist that he required repentance as a prerequisite for his physical cleansing. Evidently his teaching left some danger of ex opere operato —

2. For the translation of Tov8ala see Burkitt, Vestigia 485k 3. This passage has been repeatedly misunderstood by neglect of its polemic character, although that was recognized (but misinterpreted) by Gottstein, Traits. Against Flusser, Sect, and Betz, Proselytentaufe, it must be said that nothing in the Qumran documents implies the use by the sect of any type of im mersion other than such as are prescribed in the OT for purification. In particular, the regulations for entrance to the sect, which are twice described in the Manual of Discipline (I and VI) and once in the Damascus Document (VI.i4ff) say nothing of any special immersions, nor does Josephus mention any such rite of admission to the sect. See further Rowley, Baptism·, Benoit, Qumran 280.

THE BACKGROUND

not to say libertine — interpretation (with the polemic in the Manual of Discipline

III.4f, cf. Connolly, Didascalia VI. 22, p. 254).

Admittedly, there is no being sure that the Qumran polemic was directed against the Baptist — rites of immersion were popular at the time, as Thomas, Mouvement , has shown. Nevertheless, his importance makes the hypothesis plausible; and so does the fact that his baptism was remembered as a distinct rite, "the baptism of John" (Mk. 11.30 and parallels; Lk. 7.29; Acts 1.22; 10.37; 1 3.24 ; 18.25; which was sharply differentiated from Christian baptism (Acts 18.25; Ι9·3θ and also from ordinary Jewish immersions for purification (Mk. 7.4, Heb. 9.10 — the βαπτιστών διδαχή of Heb. 6.2 was presumably instruction as to the different sorts of baptisms then in competition; cf. Spicq, ad loc.).

The characteristics of the Baptist's rite have often been noticed : It was not some thing one could do for oneself; it had to be administered by the Baptist or one of his disciples, was perhaps public, was administered to Jews (whether or not to gentiles), may not have been repeatable, used water and probably required immersion, either required or effected repentance, was accompanied by confession of sins, effected remission of sins, demanded the performance of good works in the future, was performed as a preparation for the coming judgment or kingdom of God, and either was not connected with any teaching about the holy spirit, or supposed the spirit would be given (as a further "baptism") only at the last judgment (Acts 19.2 vs.

Mk. 1.8 and parallels; see Best, Spirit). That it made the recipient a member of a new community is often said by the critics but never by the sources — there is no evidence that all those whom John baptized became his disciples; the reports of crowds coming to be baptized suggest that the rite entailed no membership in any society, and Josephus' βαπτίσμω συνιέναι is inadequate as an excuse for supposing the contrary (vs. Flemington, Doctrine 15). What was demanded of the few who did become disciples, we do not know.

It has already been remarked that no baptism of this kind appears in the Qumran documents. Attempts to derive it from Jewish proselyte baptism are equally mistaken.

(See the distinctions made by Werblowsky, Rite ioif; Doeve, Doop ; Michaelis, Hintergrund; Beasley-Murray, Baptism 40-42; and the chronological arguments of Taylor, Beginning.) It should be added that the rule "a proselyte is as a new born child" is a legal fiction meaning that the proselyte is freed from most legal liabilities

— including the liability for transgressions — contracted in his earlier life, and also loses his prior legal claims — for example, he has no property. It does not mean — pace Werblowsky, Rite 102 — that he is reborn as a child of Israel; Daube's examples,

Reflections 51, show loss of prior legal ties, including family connections, but not acquisition of new ones. Rabbi ruled in Bikkurim 1. 4 that proselytes must continue to speak of the Patriarchs as "their fathers," not "our fathers." Though the contrary opinion of R. Judah (J. Bikkurim 1. 4 (64a), called to my attention by R. William

G. Braude) subsequently prevailed, the fact that neither the Mishnah nor the Tosephta mentions it makes almost certain — so R. Saul Lieberman advises me — that the pref

erence of it is a subsequent alteration of the law. In any event, R. Judah based his opinion on Gen. 17.5, "For I have made you the ancestor of a multitude of peoples,"

CLEMENT OF ALEXANDRIA

so he presumably considered the proselytes a different people than Israel and justified their reference to "our" fathers by the fact that they were descendants of Abraham, not of Jacob — this would have involved no change of ancestry. Moreover it seems likely that his opinion was a reply to Christian propaganda, see the use of the same verse by Paul, Romans 4. 1 7ff, and the development of this by the fathers, for instance

Aphraates, Demonstratio XI. 1 off, who is evidently answering Jewish arguments.

Since, then, the Baptist's rite can be derived neither from Pharisaic baptism of proselytes nor from Qumran, we must look elsewhere for its origin. The true source is that indicated by Mk. 11.30: it was from "Heaven," that is, from God. This was what all the people believed (ibid.), and their belief presumably came from the Baptist's own claim — that he was a prophet (Mt. 11.9 and parallels) charged by God to institute a new rite, an immersion which would remit sins (cf. Kraft, Anfange

40if).

To understand the importance of this one must realize that in Jewish law sin and impurity are different things. One may become highly impure by any number of accidents which involve no sin whatever— for example, by being present in a house when a death occurs there. Many sins, on the other hand — theft, for instance — do not render the sinner impure. Admittedly, by the Baptist's time the distinction had often been blurred. The prophets and psalmists had spoken of sins as impurities from which men could be cleansed (e.g., Ps. 51.4) ; the verb "to sin" had an intensive form which should have meant "to de-sin" but actually meant "to purify," and the water used for certain types of purification was known as "sin-water." But neither poetry nor etymology is valid in law. In law, impurity was removed by immersions (the "baptisms" of Mk. 7.4), and certain unintentional sins could be expiated by the sacrifices prescribed in the OT ; but for the rest one could only trust to the general expiatory rites of the day of atonement, to the powers of repentance, restitution, reformation, and prayer, to the atonement effected by one's death, and to the mercy of God ( Toma ' VIII. 8f). Moreover, the possibilities for sacrifice to Yahweh in Palestine had been cut down sharply under the Maccabees; the destruc tions of the temples at Gerizim, Araq el Emir, and Lachish indicate what happened to lesser shrines. By John's time the only place in the country where Jews could legally offer sacrifices was Jerusalem, and its services were expensive. To introduce into this situation a new, inexpensive, generally available, divinely authorized rite , effective for the remission of all sins, was John's great innovation.4 His warning of the coming judgment was nothing new; prophets had been predicting that for the past eight centuries. The new thing was the assurance that there was something the average man could easily do to prepare himself for the catastrophic coming of the kingdom. Therefore John was remembered not merely as a prophet, but as "more than a prophet" — as "the Baptist."

4. The Qumran sect had probably developed the doctrine that their life of obedience to the Law had the atoning power of the sacrifices (see Flusser, Sect. 229-236); but the devotion of an entire life to the observance of an ascetic law is a very different thing from the pei'formance of one, quick, easy ceremony.

THE BACKGROUND

VIII. Evidence from the Gospels that Jesus baptized Since the Baptist's role in relation to the coming kingdom is so clearly definable, it is surprising that there should be such obscurity about the role of Jesus— who is said to have been thought, in his own lifetime, the Baptist redivivus (Mk. 6.14;

8.28 and parallels; cf., however, Kraeling, Necromancy) .

Particularly surprising is the particular obscurity concerning Jesus' use of baptism.

He himself was baptized; but the synoptics say nothing of his having baptized his followers, while the Fourth Gospel contradicts itself on this point (as on so many others): it says in 3.22 that he did baptize; it refers to reports of his baptizing in 3.26 and 4.1 ; but in 4.2 it adds καίτοι ye Τησοΰς αυτός ονκ έβάπτισεν άλλ' οί μαθηται αύτοΰ. Here the accepted alternatives are either to eliminate 4.2 as a gloss (Bultmann,

Johannes 128 n4; Dodd, Fourth Gospel 31 1 and Historical Tradition 237) or to accept it as a correction: Jesus did not himself baptize, but during an early ministry in Judea

(unknown to the synoptics except for Lk. 4.44, on which cf. Acts 10.37 — above, p. 206) he permitted his disciples to go on administering (the Baptist's?) baptism, a practice which he subsequently stopped (why?) but which the disciples nevertheless resumed as soon as he was out of the way, Acts 2.38. But here the function of the Baptist's rite — els dpeaiv των αμαρτιών — is combined with the Christian formula eV I τω όνόματι Τησοΰ Χρίστου. How the disciples could baptize in Jesus' name if Jesus had stopped the practice is not easily explicable (see the wrigglings of BeasleyMurray, Baptism 7of; Flemington, Doctrine 3of, chose to ignore the problem).

These difficulties may perhaps justify a hypothesis of mistranslation. In Syriac

4.2 reads, κ\κ .Kom ^.cob ot> ko«d νΔ TL3 . It is only the medial period which prevents this from meaning, "Yet Jesus himself baptized none save his disciples."

(A \> before ,α>οπ>^λ would be desirable, but not necessary. Cf. the similarly possible mistranslation in Jn. 12.17, above, p. 157.) Variant forms of the tradition that Jesus baptized only his disciples are found in Clement (III. 196.2 iff) and later writers (Echle, Baptism 367^, and this is the meaning indicated by the context in Jn· 4.2. Jn· has been at its usual business of contrasting Jesus and the Baptist to the latter's disadvantage. It has just reported how the Pharisees heard ort Τησοΰς TrXelovas μαθητας i roiei και βαπτίζ€ΐ η Τωάννης. Here the Syriac reads

.^uoi yjo kku^co In any case the Greek should not be read as evidence that all whom the Baptist baptized became his disciples. The evangelist was talking about Jesus. He realized the ambiguity of his statement (which was to mislead Bultmann, Johannes 128 n7) and therefore went on to explain it by declaring,

"Yet Jesus baptized only his disciples ! ''—while the Baptist (it is to be understood) baptized everybody who came to him — and even so Jesus' disciples were more numerous than the Baptist's here-today-and-gone-tomorrow penitents. This pre sumable exaggeration concludes the theme begun in 3.26 and formally stated in 3.30. (See also Dodd's arguments in Historical Tradition 285^ 292, and, further, from the themes of water and spirit in Fourth Gospel 308—31 1.) [Discussion with C.R.

leaves me dubious about the above interpretation. It is possible that the author

CLEMENT OF ALEXANDRIA

(or glossator) meant, "Tet Jesus baptized only his disciples, and consequently the Pharisees were misinformed and their (suggested) plots unnecessary."]

Thus we have the statements of Jn. on one side, the silence of the synoptics on the other. To Jn.'s statements should be added the account of the footwashing in 13.1-15. The footwashing is to baptism as the feeding of the multitude is to the eucharist; for both sacraments Jn. has a chapter of theological exegesis (3 on baptism,

6 on the eucharist) and a story of a similar event which could be used by a teacher as a type of the rite. Aphraates, Demonstratio XII. 10 understood the footwashing as a miraculous baptism, by anticipation, of the twelve, into Jesus' passion. As such he contrasted it with the Baptist's rite, which only produced remission of sins by repentance, and he thus explained why the Baptist's penitents were rebaptized, with the baptism of Jesus, by the apostles. But Jn., as usual, remains enigmatic.

Synoptic evidence that Jesus baptized might be found in Mk. 1.8 and parallels (including Jn. 1.33), where the Baptist is made to declare that he can baptize only with water, but his greater successor will baptize with the holy spirit ("and with fire," QJ ; and also in Mk. 9.49 (correcting αλισθησζται to βαπτίσθήσεται, with Baarda). But Acts (1.5; 11.16) interprets the promise of baptism with spirit and fire as referring to Pentecost (Acts 2.3^17), and Jn. 7.39 indicates that the spirit was not given until after Jesus' resurrection; so it could be that the whole of the Baptist's prophecy is an anachronistic invention of the Christian polemic against his followers. (The greater successor — whose shoelaces he was not worthy to untie — can hardly have been Yahweh, who has no body, parts, or shoelaces. Contrast

Kraft, Anfange 400.) Mk. 9.49, as a prophecy by Jesus, is, rather, evidence against his practice of any such rite. That MSS c and e at Lk. 23.5 add to the Jews' charges against Jesus, et filios nostros et uxores avertit a nobis , non enim baptizantur sicut nos , is dubious evidence at best, and the first phrase is attributed to Marcion by Epiphanius

Panarion XLII.i 1.6—8 scholion 70, ed. Holl, p. 116) — but the attributed phrase is in a slightly different form and place, so might be taken as collateral evidence for the tradition.

In sum, we come back again to the self-contradictory statements of Jn. and the silence of the synoptics. Jn.'s statements might be explained as polemic, to set Jesus above the Baptist, but self-contradiction is not a common polemic device, and one's general impression of Jn. is that it misrepresents more by exaggeration than by baseless invention. On the other hand, the strength of arguments from the silence of the synoptics has been weakened considerably by the evidence we have seen for thinking

Mk. an exoteric work. Accordingly, since neither is conclusive, the contradictory evidence of Jn.'s statements and the synoptics' silence must be judged by relation to the larger problems of the history. These indicate that Jesus baptized.

Foremost among them is the obscurity, which we have been trying to penetrate, of Jesus' function in relation to the kingdom. We have seen his expected role in the final establishment, and his present role as advance agent. But we have also seen that these are inadequate, because they indicate nothing unusual for his hearers to do about his announcements — and this inadequacy is particularly glaring by contrast Avith the clear and practical function of the Baptist.

THE BACKGROUND

Moreover, against the supposition that Jesus gave up the practical rite of the Baptist and went back to the mere preaching of repentance traditional to the prophets stands the fact that the Gospels do not consistently represent him as a preacher of repentance. They contain no accounts of mass penitence produced by his preaching, and few of individuals' repentance (and those few are suspect — e.g., Lk. 19.1-10).

The message which Mk. 1.15 and parallels put into Jesus' mouth is not his, but that of the later Christian preachers : μετανοείτε και πιστεύετε εν τω ευαγγελίου (that is, the gospel about Jesus; against Taylor see Lohmeyer, and Bultmann, Geschichte,

124). Accordingly, it is plausible to suppose the same origin for the reference to repentance in Mk. 6.12 (a late addition unknown to both Mt. and Lk. ?). These two are the only references in Mk. to Christian preaching of repentance. Jn. never refers to μετάνοια or to μετανοεΐν. ζ) has the verb in its exegesis of the sign of Jonah

(Lk. 11.32 II Mt. 12.41 — evidently posterior to the resurrection) and in the woes on the Galilean cities (Lk. 10.13 || Mt. 11.21), which are thus the best evidence that Jesus occasionally did use the theme. It would be hard to believe that he did not, but it would be even harder to believe, in the face of this evidence, that he was principally a preacher of repentance.

The difficulty becomes even clearer when one remembers that against this exiguous evidence must be set not only the rarity of stories of repentance, and the rarity of any other examples of Jesus' preaching it (found only in Lk.), but also the presence of a great deal of material which represents Jesus and his followers as anything but penitents. The Baptist, who certainly did preach repentance, conducted himself as a penitent; Jesus, by contrast, "came eating and drinking" (Lk. 7.34 || Mt.

1 1. 19). The followers of the Baptist and those of the Pharisees fasted; Jesus' followers did not, and Jesus justified this by comparing them to the members of a bridal party (Mk. 2.i8ff and parallels). He justified their laxity in observance of the Law by comparing them to the companions of David (Mk. 2.25 and parallels). He forgave sins freely, without demanding repentance (Mk. 2.5 and parallels). He blessed not the penitent and the fasting, but the poor and the hungry (Lk. 6.2ofif || Mt. 5·3ίΓ).

And so on — the theme is familiar and need not be developed at length : The Gospels simply do not represent Jesus as principally a preacher of repentance; in this respect he differed fundamentally from the Baptist, and the difference was noticed and criticized in his own time. It now has to be accounted for.

Ever since Dodd's Parables this difference has customarily been accounted for by saying that for the Baptist the kingdom is still in the future, for Jesus the kingdom is here. This position is now familiar and does not need exposition, but development.

If the kingdom is simply the realm of obedience to God, then it should come whenever

God is obeyed. Any prophet preaching repentance will restore it insofar as his preaching is successful, and repentance will be the primary step in its restoration.

Therefore the emphasis placed by the Gospels on the fact that Jesus was not merely a prophet (Mk. 8.28f and parallels; Lk. 7.26ff || Mt. n.gff; Jn. 4.19-26; 7·4θίΓ; etc.), and the inconspicuous role of repentance in his teaching and his followers' practice, indicate some different notion of the way in which the kingdom is piesent and the consequence of its presence. This notion presumably concerns some special

21 1

CLEMENT OF ALEXANDRIA

function of Jesus in relation to the present kingdom, and also, in the light of the Baptist's career, some special, practical, probably ritual, response which could be made by Jesus' followers to his announcement of the kingdom's presence. Now a few elements in the synoptics represent the kingdom as already — in some respect — attainable, and the disciples as already in it:

Mk. 2.19

The members of the wedding party cannot be made to fast.

David gave those who were with him the holy bread.

2.26

The disciples already need not observe the sabbath.

The twelve are empowered, already, to exorcise (again 6.7).

3·ϊ5 3-35 Those who do the will of God are already Jesus' kin.

The mystery of the kingdom has been given to the disciples.

The kingdom is already growing like a mustard seed.

The disciples already need not observe the purity laws.

The transfiguration.

4.1 1 4.3°ff 7.2ff g.2ff 9.i8ff,28 The disciples practiced exorcism (9.38, so did others, in Jesus' name). io.29f The disciples will be rewarded now — but with persecutions.

The eucharist.

1 4»22ff Q (The references are to Lk.; generally I follow Manson, Sayings.) The least in the kingdom is greater than Jn.

7.28

The disciple should let the dead bury the dead (he is already alive).

10.9

9.60

The disciples are, already, to heal the sick and announce the kingdom; Mt. adds, raise the dead!

He that heareth you heareth me, etc.

10. 16

10.21

The Father has revealed these things to babes (and hidden them from the wise).

10.22

The Father is now revealed by the Son.

Whoever asks receives now, etc.

1 1. 10

11.52

Those who were entering (the kingdom, Mt. 23.14) were hindered 12.3 by the lawyers.

The things told the disciples now in secret are someday to be proclaimed openly.

12. 31

Seek first the kingdom, and the good things of this world will be added.

The kingdom is already spreading like leaven.

The law and the prophets were until John, thenceforth (i.e., 13.21

16. 16 now), everyone forces his way into the kingdom.

To him that hath shall be given.

19.26

Mt. 1 1.28

Come unto me and I will give you rest now.

II.29f Take my yoke {sc. of the kingdom) now.

The field with the treasure can be purchased now.

THE BACKGROUND

13.46 The pearl can be purchased now.

13.52 A scribe can already be a disciple of the kingdom.

16.19 The keys of the kingdom shall be given to Peter now (so that what he binds on earth shall be also at once bound in heaven).

18.20 Where two or three are gathered in my name, I am among them now.

Lk. 10.17

The demons are already subject to the disciples, whose names are already written in heaven.

The repentant son is admitted at once to the feast.

The kingdom is in your power (Griffiths, Within).

17.21

Some of these elements are much clearer and more conclusive than others, but the clear ones indicate how the others should be interpreted. Many different sorts and strata of material are represented, but this shows that the notion is not a peculi arity of any one strand of the tradition. Its wide distribution argues an early, common source, and the argument is strengthened by the present union of the disciples with Jesus in Jn. (15. iff, etc.), and the location of the believers in the kingdom by Paul

(Col. 1. 1 3) and by the Apocalypse (1.6, etc.). All in all, it would seem that Jesus somehow enabled at least some of his followers to enter the kingdom forthwith, and to enter it in some special fashion other than that of mere repentance and obedience to God — some fashion which would make them greater than the Baptist (who was presumably penitent and obedient), would exempt them from the Law, give them power over demons and diseases, and admit them at once to the feast. This admission was the special function of Jesus, the Messiah — the function which the Baptist, though more than a prophet, could not perform.

Now the recognized means of preparation for admission to the kingdom — recog nized by Jesus himself, since he had used it — was baptism, and in the earliest Christian material we find baptism the means of admission to the Church, which is the kingdom present on earth. It is therefore reasonable to assume that Jesus effected the admission of his chosen followers to the kingdom by some sort of baptism. On the other hand, the Baptist's baptism evidently did not admit the recipient to the kingdom, so Jesus' baptism must have differed from it. First and foremost it differed in being not only private, but secret. Therefore, attempts to determine its other differences are necessarily speculative.

IX. Baptism according to Paul As a check on speculation, it is helpful to compare the Baptist's baptism, which dates from about a.d. 25, with the Christian baptism as we find it in the earliest

Christian documents — Paul's letters of about a.d. 50. There is admittedly some doubt as to when Paul is talking about baptism ; here we shall follow the example of

CLEMENT OF ALEXANDRIA

Braumann, Taufverkundigung , and confine ourselves to passages where the reference is indubitable. Of these the most important are the following:

Rom. 6.gff: Those of us who were baptized into Messiah Jesus were baptized into his death. That is to say, we were buried with him, through the baptism into death, in order that, just as Messiah was raised from the dead through the glory of the Father, thus we too should live a new life. For if we have been united with him in a death like his, let us be so, too, in a resurrection like his. For we know that the man we once were was crucified with him, in order that the body which belonged to sin might be made ineffective, so that we should no longer be slaves to sin. For when a man dies he is no longer answerable for his sins. And if we died with Messiah, we believe that we shall also continue to live with him, knowing that Messiah, having been raised from the dead, will never again die.

I Cor. 12.12/ : For just as the body is one, but has many members ... so also is the Messiah.

And thus we all were baptized with one spirit to constitute one body — whether Jews or Greeks, whether slaves or freemen — and we were all given one spirit to drink.

Gal. $.26ff: For all of you {formerly gentiles) are sons of God through faith in Messiah

Jesus. For as many of you as were baptized into Messiah have been clothed with Messiah.

In him there is neither Jew nor Greek, there is neither slave nor freeman, there is no male and female, for you all are one in Messiah Jesus. And if you are Messiah's, then you are the seed of Abraham and heirs of the promise {made to Abraham by God).

Col. 2.gff: {In the Messiah) all the fullness of the divine dwells bodily. And you are fulfilled in him, since he is the head of every cosmic power and authority. In him you have also been circumcized, not with the physical image of circumcision, but with the stripping off of the body of flesh, with the circumcision of the Messiah, having been buried with him in baptism, in which you have also been resurrected together with him through faith in the working of God who raised him from the dead. Thus, when you were dead in your sins and in the foreskin of your flesh, God brought you to life together with him. Having forgiven us all our sins, he {the Messiah) has cancelled the bond with the legal demands which was against us and has set it aside, having nailed it to the cross. Having stripped off the cosmic powers and authorities he has made a public spectacle of them and led them, by means of the cross, as captives in his triumphal procession. Therefore let no man sit in judgment on you about food and drink, or in a matter of festival or new moon or sabbath, which are a shadow of the things to come, whereas the substance is of the Messiah. Do not let yourselves be condemned by anyone set on self-abasement and worship of angels, things he saw going in {to the heavens),5 someone puffed up to no purpose by carnal imaginations, and not holding to the head from which all the body, nourished and knit together throughout all its joints and ligaments, grows the growth of God. If, dying with Messiah, you left behind the elemental spirits of the world, why do you live as if still in the world? ... If you have been raised from the dead with the Messiah, seek the things above, where the Messiah is, sitting to the right of God. Fix your mind on the things above, not those on earth. For you have died, and your life has been hidden with the Messiah in God. When the Messiah, our life, shall be revealed, then you too will be revealed with him in glory.

Finally, we must add I. Cor . 10. 1-4, although it does not speak directly of Christian baptism :

5. See my Observations, 156-157.

THE BACKGROUND

I would not have you ignorant, brethren, of the fact that our fathers were all under the cloud and all went through the sea and were all baptized into Moses in the cloud and in the sea, and all ate the same spiritual food and all drank the same spiritual drink, for they drank from the spiritual rock which followed them, and the rock was the Messiah.6

The above passages show at a glance the immense difference between the baptism of the Baptist and that of Paul. The former was analogous to earlier biblical immer sions except that it was specially instituted by God through a new prophet and it removed sin, whereas they had removed impurity. The baptism of Paul, on the contrary, is essentially a means of uniting with the Messiah. Since the Messiah and the spirit are so closely related as to be practically identical— Paul once explicitly identifies them (II Cor. 3.17, "The Lord is the spirit," cf. I Cor. 15.45)- — the union is conceived as possession by a spirit. The spirit dwells in the baptized (οίκεΐ eV,

Rom. 8.9, 11 bis; I Cor. 3.16; cf. I Cor. 6.19 and Philo, De somniis 1 . 1 48Γ, cited above, p. 171), and acts through them (I Cor. 12 and passim ; notice especially the spirit's "speaking" through the possessed — Rom. 8.26, στεναγμοϊς άλαλητοίς, cf. I Cor. 2.13 and all of ch. 12 — a phenomenon often observed in schizophrenia; cf. Mk.

1.24; 5.7; 1 3. 1 1 ; Philostratus, Vita Apollonii III. 38). Thus the body of the possessed

Christian is in effect a part of the body of the Messiah, the spirit which lives in each

Christian and acts through him. All Christians together constitute the whole body of the Messiah, of which each individual body is a member (that is, an organ — a hand, or a foot, or whatever). As I Cor. 12.13 says, "we were baptized with one spirit to constitute (et? ) one body." (Cf. Plutarch, De Iside 73 (380c): The soul of Typhon is divided among the various Typhonic animals.)

This central concept Paul develops in different ways to meet the needs of different situations. In Romans 6 , where he is protesting against a libertine interpretation of his teachings, he argues that union with the Messiah involves participation in his death and resurrection; since the death was a death to sin, the new life is a life to God, from which sin is necessarily excluded. (Rom. 6.1 if. The future in verse 5 is hortatory; that the resurrected life has already begun is clear from the context.

The future is used in verse 8 because the present resurrected life will continue; cf.

Moule, Idiom-Book 23; Sanday-Headlam, Romans 154-155 and n. on 6.8.) We shall see later that Paul's notion of baptism as death and resurrection may have resulted

6. Here it is a mistake to speak as Lietzmann ( Korinther , ad loc.), does, of topology. The concluding explanation ("for . . . Messiah") must have been needed to explain why the food and drink were "the same." "The same" is repeated for emphasis. The same as what? The same as what we Christians now eat and drink, "for . . . the rock was the Messiah." This is the point of the whole passage: Baptism and communion will not make you wholly immune to the consequence of eating things offered to idols, for they did not produce such immunity in the generation of the exodus. I his argument requires that baptism into Moses should be "the same" thing as baptism into Christ — not a mere type of it and so it would be for any Christian who had formerly been a reader of Philo and knew that Moses had been, like Jesus, an incarnation of the Logos. Therefore Paul does not have to explain this as he does his notion about the rock, which depends more on Palestinian Jewish traditions (Alio, I Cor., ad loc.). This

shows he expected most members of his Corinthian church to be familiar with exegesis of the Philonic type on the story of the exodus. This is evidence in favor of Goodenough's hypothesis (1.23—29) that such exegesis was once widespread in Greco-Roman Jewry.

CLEMENT OF ALEXANDRIA

from other causes as well as from the notion of union with Jesus. Here let it be noted only as a development both so important and so unlikely as to deserve attention.

In I Cor. 12 , where Paul is protesting against the arrogance of those who claimed special spiritual gifts, he argues that union with the Messiah implies that all Christians have the same spirit and are therefore parts of the same body and mutually dependent.

In Galatians where he is protesting against the pretensions of those who claimed to keep the Jewish law, he argues that union with the Messiah implies that all Chris tians are essentially identical, for all are the Messiah, the true seed of Abraham, the true heir of the promise. (Here his use of the metaphor "you have clothed yourselves in Messiah" is probably an allegorization of the clothing which followed baptism, and suggests that the baptism was naked. See above, pp. 1 7 5f ; and, further, I Cor.

1 5*53? where the baptismal reference was already recognized by Odes of Solomon 15.8; also I Cor. 15.44-49, cf. Robinson, Hymn 62 and 72ff, esp. 77-78; also II Cor.

3. 1 7f ; 5.2.) Finally, in Colossians 2-3, where he is protesting against the introduction of some cult of the cosmic powers (which conceived of them as supernatural beings to be honored or placated by observance of rules of the Mosaic Law), he argues that union with the Messiah involves participation in his nature, his death, and his resurrection. By nature he was superior to all the cosmic powers, by death he stripped them off and subjected them, and by resurrection he ascended above them, and was hid in God, where the Christians are hidden with him until the end. Therefore they should not subject themselves to the laws of inferior beings. Notice the recurrence of the stripping-then-clothing motif in 2.11,15 and 3.9. Here it is connected with the legend of the descent of a supernatural being to the lower world, his assumption from it of a physical body or some other sort of disguise, and his return to the heavens, stripping off the disguise and resuming his true form (cf. Bousset, Himmelsreisse

1 39— 1 4 1 , 233 and n2). This legend was ancient and widespread in Orphic and Iranian mythology (Bidez, Ecoles Puech, Ou 30 iff; Cumont, Religions 282f 069), was popularized by Plato, and about Paul's time found expressions as different as CH I.12-26, the Naassene hymn in Hippolytus, Philosophumena V. 10, and The Hymn of the Soul (cf. Preuschen, Hymnen 6 if). Compare also the Prayer of Joseph (James,

Lost Apocrypha 2 iff; Nock, review of Schoeps 584 — Nock's discussion of the descent legend, 584-590, is vitiated by a scholarly concern for details which prevented him from recognizing the common pattern variously adapted in the various stories).

The legend was already used by Christians as an interpretation of Jesus' work in the pre-Pauline hymn quoted by Paul in Phil. 2.6ff. We have also remarked its use by Paul, Simon Magus, Mark (?), and the Ascension of Isaiah (see above, p. 201).

It is also basic to the thought of Jn. (1.9-11; 16.28; 19. 11) and Hebrews (i.2f), and is epidemic in gnosticism and neoplatonism (Bousset, Hauptprobleme 36 iff).

X. Elements derived from Jesus in Pauline baptism Some of these Pauline interpretations and applications of baptism are recognizably secondary. The notion that identification with Jesus involves participation in Jesus'

THE BACKGROUND

death and resurrection is obviously later than those events. (Contrast the Ebionite tradition that the essential for salvation is not the sacrifice of Jesus, but baptism:

Schoeps, Judenchristentum 69, 84b This at least could have been the teaching of Jesus.)

Paul's deduction of mutual dependence from the analogy of the body looks like moralization produced to meet the needs of a developing community. Similarly, the argument that all members of the Messiah must be parts of the seed of Abraham can have been pressed only in a community where Jewish snobbery and gentile emulation were a problem. Finally, the identification of baptism into Jesus with baptism into Moses is a consequence of the Philonic type of logos theory, which does not seem to have been common in Jesus' circle. All these, therefore, are probably posterior developments.

The case is different for the essential Pauline notion that baptism results in the possession of the baptized by the spirit of Jesus. This reflects demonological beliefs which appear in the exorcisms in Mk. It may have its background in the same

Palestinian milieu. Compare the way the Palestinian apocalyptic writers "explain inspiration ultimately in terms of possession," Russell, Apocalyptic 175. (Innumerable attempts have been made to find the origin of the Pauline notion in paganism, especially in the mysteries; these are reviewed and rejected by Wagner, Problem. The rejection was evidently predetermined, but nevertheless seems to me justified. Cf.

Warnach, Tauflehre.) Moreover, there are a number of reasons for thinking that Pauline baptism came not only generally from Palestine, but specifically from Jesus: A. It was essentially a means of uniting with Jesus. B. This union was effected by the spirit, which Jesus had. C. The closest analogies to the rite are found in magical material, and there is considerable evidence that Jesus practiced magic. D. Paul's rite was soon and widely connected with ascent to the heavens, with which Jesus was also credited.

E. It freed the recipient from the obligations of the law, from which Jesus' disciples were freed. Let us examine these points in order.

A. The rite was a means of uniting with Jesus Because it was a means of uniting with Jesus, Pauline baptism was radically unlike any rite known from Palestinian Jewish tradition except the eucharist. If Goodenough be correct in supposing (1.6, etc.) that some rite was devised or inter preted to symbolize or effect union with the Logos, this provides an analogy. But the evidence is inconclusive and the analogy — if any— remote. Ί he demonological character of Paul's concept indicates that his rite did not come from the philosophers of Alexandria, but from the magicians of Palestine. Its magical character was long ago remarked by Dieterich, Mithrasliturgie 1 78. On the prevalence of magic in Palestinian Jewry see Schiirer, Geschichte III.407ff and Lieberman, Greek 91 — 1 1 4.

The evidence is rich, from Jewish and pagan sources alike. Spells in which the magician identifies himself with a spirit are plentiful in the magical papyri. A good Jewish

CLEMENT OF ALEXANDRIA

example comes from The Sacred Hidden Book of Moses called Eighth or Holy ( PGM XIII.343ff)j lines 783^ :

"And Thou, lord of life, ruler of the heavens and the earth and all those dwelling in them, whose righteousness is not turned aside, whose glorious name the Muses hymn, whom the eight guards escort, Η, Ω , Χω, Χουχ, Νουν, Ναυνι, Άμουν, Άμαννί, who hast the unerring truth: thy name and thy spirit rest upon the good; enter my mind and my thoughts for the whole time of my life and perform for me all the desires of my soul, for Thou art I and I am Thou. Whatever I say must happen, for I have thy name as sole amulet in my heart, and no disturbance of the flesh shall overpower me, no spirit shall oppose me, no demon nor visitation nor any of the evil beings of Hades, through thy name, which I have in my soul" (cf. Rom. 8.38f; Apoc. 3.12;

14. i ; 22.4).

There is another ritual means of uniting with Jesus — the eucharist — which is even less compatible with the material commonly taken to represent "normative"

Judaism (eat my body! drink my blood!), but which pretty certainly was introduced by Jesus and exemplifies the same sort of magical thought and practice. Cf. DMP

XV.1-19:

One mingles various ingredients in a cup of wine and says over it an invocation :

"I am he of Abydos . . , as to which the blood of Osiris bore witness . . . when it

(the blood) was poured into this cup, this wine. Give it, blood of Osiris (that?) he (?) gave to Isis to make her feel love in her heart for him . . . give it, the blood of N. born of N. ... to N. born of N. in this cup, this bowl of wine, today, to cause her to feel a love for him in her heart, the love that Isis felt for Osiris."

This type of magical procedure is standard; cf. PGM no. VII, lines 643ff, which comes even closer to "this is my body"; the wine is made into the flesh ( σπλάγχνα ) of Osiris and Iao (= Yahweh). On the relation of the eucharist to the mysteries see

Nock, EGC and Mysteries. As in the case of baptism, the eucharist's obvious incom patibility with supposedly "normative" Judaism has led to repeated efforts to derive it from the mysteries. It is a fact that Dionysiae rites celebrating the god's gift of wine were practiced around Galilee in Jesus' time; see my article "On the Wine God in Palestine," in the forthcoming Festschrift for S. Baron. These rites probably derived from a native Syro-Palestinian cult of a wine god; the Greek myth associated with them referred to the wine as "blood"; the god was said to enter those who drank the wine; Jn. 2. i— 1 1 was modelled on this myth; there are other traces of its influence in Jn. and the influence of the cult of the wine god can be traced in many elements of Palestinian Judaism from Genesis on. But the influence of the mysteries and the influence of magic are not mutually exclusive— indeed, magicians were commonly said to have established mysteries (Burkert, ΓΟΗΣ 39f;

Lucian, Alexander 38; etc.), and certainly drew material from them (notably for the figures of Orpheus, Hecate, and Selene: Nilsson, Zauberpapyri 67, 7 iff). Moreover, the Dionysiae myth is not enough to explain Jesus' institution of the eucharist: the myth tells of what a god once did, but provides no excuse for a man's doing it.

That a man should undertake to identify his own blood with wine and give it to his followers to drink in order to unite them with himself— this goes far beyond the

THE BACKGROUND

mysteries; its only close parallels are in magic. And there is plenty of evidence that magic was common in Palestine; e.g., B.Berakot 53a (end); Sifre Devarim 26; Sot ah

IX. 13 ; J.Kiddushin IV.11 (66c); Gressmann, Aufgaben 11— 15.

Thus early Christianity has two rites for uniting the believer with Jesus. Both derive from the same type of magical practice, both show a similar break with traditional Judaism, and both must have been introduced within half a dozen years, if we are to accept Acts' stories of baptisms in the first days of the Jerusalem community and of the baptism of Paul. Of these two rites, one was certainly introduced by Jesus; the presumption that the other one also came from him is strong. If it be accepted, it enables us to understand in a new sense a number of sayings in which the unity of the disciples with Jesus is implied, e.g., Lk. io.i6f and parallel: He that heareth you heareth me.

B. The union was effected, by the spirit In Pauline baptism the union with Jesus was effected by the spirit. OT immersions have nothing to do with the spirit; neither does proselyte baptism. The Manual of Discipline III^fT required cleansing by the spirit as a prerequisite for valid immersion for purity; so the immersion probably was not thought to give the spirit. The Baptist's baptism, too, is contrasted with Christian on the ground that it was only with water, not with the spirit. (So Mk. 1.8 and parallels; Acts I9.2ff. On 18.24-28 see

Flemington, Doctrine 40.) Whence, then, did the spirit come into connection with baptism? It first appears at the baptism of Jesus. The variety of the traditions which agree on this point suggests that some historical fact may lie behind it. Compare Mk. 1.1 1 and parallels

(presumably QJ ; Jn. 1.33; Ebionite Gospel; Gospel according to the Hebrews; Col. 1.19

(?— see Miinderlein, Erwahlung) . Jn. 1.33 is particularly interesting for our purpose.

Though suppressing Jesus' baptism (because that subordinated him to the Baptist) it implies a connection between the descent of the spirit on Jesus and his power to give it to others. And both friends and enemies agreed that Jesus had a spirit — the only question was, which ? The spirit possessed him and drove him into the wilderness to be tempted of the devil — another spirit (Mk. 1.10 and parallels). He made his reputation by casting out spirits (Mk. i.23ff,27; 3.1 1; 5-2ff; 7.25; 9. 1 7fT) . He was accused of doing so by Beelzebub or some unclean spirit, but claimed that he did so by the holy spirit (Mk. 3.22IT) ; he gave his disciples power over spirits (Mk. 6.7) and assured them that the holy spirit would speak through them (as the spirits in the demoniacs spoke through them — Mk. 1.24; 5.7; 13. 11). This Markan material is further developed in the other synoptics (e.g., Mt. 10.25; Lk. io.i7ff) and sup ported by apparently independent traditions in Jn. (e.g., 8.48; io.2of — for the magical background of this see Dieterich, Mithras liturgie 1 1 7) and the Gospel according to the Hebrews. (See, further, Samain, Magie 456ff.) Since spirits played such an

important role in the life of Jesus, and since baptism had been the occasion when he — unlike most of those whom John baptized — was seized by a spirit, it is not

CLEMENT OF ALEXANDRIA

implausible to suppose that he was the one who transformed baptism into a regular means of his followers' getting a spirit. (The tradition in Jn. and Acts that the spirit was not given until after Jesus' death will be discussed below.)

C. The rite was magical The magical character of Pauline baptism also points back to Jesus. We have already seen (above, in A) that it jibes with the magical character of the eucharist, which Jesus instituted. However, the use of the term "magical" is sure to occasion misunderstanding and requires justification.

I. THE TERM AND THE FACTS

In the Roman Empire the practice of magic was a criminal offense (Paulus, Sententiae V.23. 14—18), and "magician" was therefore a term of abuse. It still is, but the connotation has changed: now it is primarily fraud; then it was social sub version (MacMullen, Enemies I24ff). The efficacy of magic was almost universally believed, and the magician was conceived as a man who, by acquiring supernatural powers, had become a potential danger to the established authorities and the order they sought to maintain. Consequently magic was widely practiced, but rarely admitted. For Judaism, a further limiting factor was the dogma that there was no god save Yahweh. But this did not lead to the denial of the efficacy of pagan magic; that was a matter of common knowledge, not to be denied. Nor did it prevent

Jews from using the same magical practices as pagans; on the contrary, they were famous as magicians (Josephus, AJ VIII. 46). The Sepher ha-Razim, newly discovered by Margalioth, shows how, as late as the fourth or fifth century, a Jew steeped in the OT and thoroughly at home in the poetry of the synagogue could still compose a magicians' handbook, listing pagan deities and Christ among the angels of the lower heavens, prescribing the prayers and sacrifices to be offered them in magical ceremonies (among the prayers, an invocation of Helios in transliterated Greek), and concluding, on reaching the seventh heaven, with a celebration of Yahweh as the sole (that is, supreme) god. The more scrupulous Jews distinguished their marvels

— as performed by the power of the supreme god or of pure spirits— from those of the pagans, whose gods were demons and whose spirits, impure. Thus Rabbi Akiba, complaining of his own ill success in magic, said: "When a man fasts in order that an unclean spirit should rest upon him, the unclean spirit does so. A fortiori, therefore, when a man fasts in order that a pure spirit should rest upon him, it should do so.

But what can I do, since our iniquities are the cause of our difficulties, as it is said, ' For your iniquities were dividing you from your God,'" B. Sanhedrin 65b end. (The context leaves no doubt of the magical reference; it goes on to report that Rabba created a homunculus — the implication being that the holy spirit rested upon him and communicated to him its creative power.) Of course neither Akiba nor Rabba is represented in the Talmuds as a magician. "Magician," as we said, was a term of abuse.

THE BACKGROUND

Given this state of affairs, it goes without saying that Jesus is not represented by the Gospels as a magician. For the Gospels he is the Son of God in disguise.

But were his practices those of contemporary magic ? That he should be represented as a supernatural being is the first suspicious item, for this was a common claim of magicians and result of magical operations. Thus in the "Mithras Liturgy" the magician begins with a prayer that the supreme being will "breathe into me the holy spirit," and then goes on to declare, "I am the Son." ( PGM IV. 487-535; cf.

Mk. 1. 1 if.) Again, the letter of "Nephotes to Psammetichus" {ibid., 154-221), begins with directions for uniting oneself with the sun, as follows :

At any dawn you wish, when it is the third day of the moon, going to the roof of a high building, spread on the earthen roof a clean sheet (σινδόνι,ον — see above, pp. 176!). Do this with a mystagogue. Then you yourself, wearing a wreath of black ivy, after eleven o'clock, when the sun is in the midst of the heaven, lie down naked < γυμνός — ibid.y on the sheet,· looking upward, and order that your eyes be covered with a black band. Then, wrapping yourself up like a mummy, closing your eyes and keeping your face toward the sun, begin the following prayer: "Powerful Typhon, sovereign and ruler of the realm above, God of gods, King {αβεραμενθωον formula), thou who scatterest the darkness, bringer of thunder, stormy one, who dazzlest the night, who breathest warmth into the soul, shaker of rocks, earthquake-destroyer of walls, God of foaming waves and mover of the deep ' Ιωερβηταυταυ ιμηνι !, I am he who searched through the whole world with thee and found the great Osiris, whom I brought to thee a prisoner. I am he who fought as thine ally with the gods (other texts: against the gods). I am he who locked the double doors of heaven <(Mt. 16.19) and put to sleep the invisible dragon, who stayed the sea, the tides, the streams of the rivers until thou mightest subdue this realm. I, thy soldier <(II Cor. 10.4), have been defeated by the <( astral) gods <1 Cor. 2.8); I have been cast down because of vain wrath. Raise up,

I beseech thee, thy friend, I entreat thee, and do not cast me on the earth, O King of gods αεμιναεβαρωθερρεθωραβεανιμεα ! Fill me with power, I beseech thee, and grant me this grace, that, when I shall order one of those gods to come, he shall at my spells come and appear to me quickly ναι,νεβασανατΓτατονεατττονμηνωφαεσμ'ηττατττονμτινωφ· αεσιμη' τραναπτι' πενχρη' τραυαρα' πτονμηφ ' μονραι' ανχουχαφατττα ' μονρσα' αραμει' Ιαω . . . Ιαω αηι αι/ααη"

When you say these things thrice the following sign of your union <with the god> will occur, but you, armed by your magic soul, should not be terrified. For a sea hawk, flying down, will strike you with his wings on your body <Mk. 1 . 1 o and parallels), by this very sign indica ting that you should arise. You, therefore, arise, clothe yourself in white garments, and burn uncut frankincense in drops on an earthenware incense altar, saying as follows : " I have been united with thy sacred form <11 Cor. 3.18; Phil. 2.6). I have been empowered by thy sacred name <Acts 3.16). I have received the effluence of thy goodness, Lord, God of gods, King,

Daimon, a θθοικν θονθουϊ ταν αντί' λαω απτατω." When you have done this, descend, having attained that nature, equal to the God's < Ισόθεος φυσις — cf. Phil. 2.6; Jn. 5.18), which is effected by this ritual union.7

It has seemed worthwhile to translate this section as a whole, not only because of the ritual parallels to the story in the longer text, but also to show that magic

7. The next five words in the Greek text are to be taken with σκεφις, which follows them.

CLEMENT OF ALEXANDRIA

does not necessarily compel the gods,8 but is capable of prayer, of individual devotion to the deity, and of considerable religious feeling. Also, besides the parallels pointed out in the text, the main notion of the passage — that by union with the ruler of the gods the initiate can attain superiority to the astral deities — is precisely that of Paul in Col. 2. 8-3. 4, discussed above.

After these asides we return to the point — that Jesus was, for the authors of the Gospels, a supernatural being in human form does not prove their thought about him was not magical. On the contrary, it suggests that it was magical. Many magical operations were designed to produce such incarnate deities; if Jesus could be made the Son of God by having the spirit descend upon him after a ritual purification, so could other magicians. That Jesus did believe he owed his powers to the possession of such a spirit is strongly suggested by the story of his dying cry, "My god, my god, why hast thou forsaken me?" (Mk. 15.34 and parallels; cf. Gospel of Peter 19, which reads "my dynamis ," equivalent to "my daimon.") From Mk. 14.50 and 15.40 it seems unlikely that any of Jesus' disciples was on hand to hear what — if anything

— he actually said. The reported cry, therefore (Ps. 22.2) is their notion of what he should have said — an expression at once of messianic hope and magical Christology.

It is plausible to suppose that the beliefs of disciples reflect those of their master.

2. THE QUESTION OF SPELLS

Since the Gospels represent Jesus as the Son of God, they credit him with the power to perform his miracles immediately. Thus he does not use charms, magical formulas, or special rituals, and this might be thought to distinguish him sharply from the magician, who is supposed to have used on every occasion the elaborate ceremonies of the magical papyri. This may have been the belief of the evangelists;

Arnobius, Adversus Nationes I-43f, and Philostratus, Vita Apollonii VI 1. 38 end, uses this argument to prove their heroes were not magicians. But many of the ceremonies in the magical papyri are initiations — means of getting a spirit. Once one has a spirit, no such rites are necessary. Thus PGM 1. 97-1 94, after describing at length the ceremonies by which one can get "the Lord of the air" (Eph. 2.2) as a familiar, concludes (lines 1 76fT) : "When you die he will prepare your body for burial, as befits a god, but, taking up your spirit, he will lead it into the air, with himself

<(I Thess. 4.17), for an aerial spirit united with a mighty familiar will not go into Hades {Acts 2.2 7ff)j, for to such a one all things are subordinate (\ Cor. 15.27).

Now when you want him to do something, say into the air only his name and ' Come,' and you will see him, and standing near you. Then say to him, 'Do thus-and-so,'

8. The assertion that the magician attempts to compel the gods, the religious man to entreat them, is so common a commonplace that it seems worthwhile to quote the explicit denial of this by Iamblichus, De mysteriis III. 18 end (ed. des Places, Paris, 1966) : το μεν οΰν θεόν η δαίμονα η άγγελον είναι τον άποτελοΰντα τα κρείττονα έργα συγχώρησε ιεν αν τις' ου μην ετι γε δίδομεν ο συ προσερριφας ώς ό μολογ ου μεν ον, ότι δι' ημών ελκόμενος άνάγκαις ταίς της κλησεως ταυτα επιτελε I. κρείττων γαρ ανάγκης εστιν 6 θεός και πας ό συναπτόμενος αύτω των κρειττόνων χορός, ου της εξ ανθρώπων επαγομενης μόνον, αλλά, και όση τον κόσμον κατείληφεν. κ.τ.λ. It must be added, however, that this philosophic protest proves the other opinion was widespread.

THE BACKGROUND

and he will do it at once and, when it is done, will say to you, cWhat else do you want, for I haste to the heaven?' And if you do not have anything to command at once, say to him, 'Go, sir,' and he will go <Mt. 8.9). Accordingly, this god will be seen only by you, nor will any hear his voice save you only {Acts 22.9)." Similar notions will be found in PGM IV. 208 iff and elsewhere.

Of course there is no more historical probability that "the Lord of the air" ever came down to answer a magician than there is that "the holy spirit" ever descended upon Jesus; but that a magician who believed he had "the Lord of the air" could perform miracles expeditiously is no less likely than that Jesus could do so, as the result of his similar belief. Hence it is clear that the absence of elaborate magical formulas from the reports of Jesus' miracles is no evidence as to whether or not he was a magician.

3. MINOR MAGICAL TRAITS OF THE MIRACLE STORIES

To make up for the absence of elaborate spells and rituals, the miracle stories in the Gospels show a great many of the minor traits of magical procedures. These were studied by Bonner, Technique 1 7 1 fT, who cited magical parallels for curing by touch, manipulation, looking upward, sighing or groaning (especially στεναγμός), the use of Aramaic phrases in a Greek context, the use of εμβριμάομ at (

1 75ff) and of ταράσσω (177). Eitrem, Demonology , added evidence for the traits noted by Bonner and himself noted further traits: the anointing with a salve compounded with spittle

(47), the emphasis on the use of the hand in touching, etc. (35; cf. the longer text

1 1 1.3-4), and especially the touching of the tongue (48), the use of "the finger of God" (34), the use of ετητάσσειν in exorcisms, and the prohibition of the demons' return (26f), the use of φιμοΰν (3of), the anger at the demons (4 if), the commands that the patients or bystanders keep the matter secret (47), the requirement that the petitioner have faith (47), and the instruction to the disciples to pray and fast before exorcisms (38). These, as Eitrem remarked, "do not seem to fit well into the picture of a wonderworking Messianic Soter" (49). Nor do they exhaust the material.

The requirement of threeor seven-day preparatory periods is frequent in PGM — e.g., IV. 1 100 (three days); IV. 26, 53, 734-735; XIII.671 (all seven days); cf. Mk.

9.2; the longer text III. 6-7; Mk. 16.1. The use of a σινδών, usually over the naked body, in magical initiations — and particularly as a costume for boys who are to serve as mediums — is a frequent and striking parallel to Mk. 14.51 and the longer text III. 8 (cf. Eusebius, Praeparatio Euangelica V.9.6; PGM 111.305,70684 IV.88f,

1708)3095; DMP III. 1 2f; XXVIII. 6; XXIX. 23). Requiring the demon to tell his name is a familiar piece of magical technique (Mk. 5.9; PGM I.160; IV. 3038;

Lucian, Philopseudes 16) — and so on; the list could easily be lengthened.

Admittedly, there are traces of a tendency in the Gospels to increase the magical traits in the stories: Eitrem, Demonology 29, notes especially the way Mk. i.298fhas been made an exorcism by Lk. 4.39; and he thinks the notion of binding, especially

by Satan, in Lk. 13.16 (p. 37) and the introduction of anointing with oil into the charges to the apostles (38) are later developments. One might add that the Johannine

CLEMENT OF ALEXANDRIA

editor of the Lazarus story has built it up with magical traits: Jesus' prefatory declaration in 1 1.25 is like many prefaces in the magical papyri (e.g., PGM V. i45ff, εγώ είμι η αλήθεια, cf. Jn. 14.6), and his prayer before the raising (

1 1 .4 1 f ) has a parallel in PGM IV.Io6of, ευχαριστώ σοι, κύριε . . . επακουων μοι επί τον της ζωής μου χρόνον. So not even the magical traits in the Gospels can be taken as certainly primitive.

However, some of them must be : they are too many to be got rid of entirely. And besides, their presence is stronger evidence than the absence of more elaborate magical material. Quite apart from the prudential and theological motivation pointed out above, the Gospels were exoteric works and we should not expect them to describe

Jesus' magical practices. Moreover, the tendency in the tradition to increase magical traits is more than matched by the tendency to diminish them — note the omission of both saliva miracles (Mk. 7.32IT; 8.22ff) by both Mt. and Lk. This latter tendency probably reflects an apologetic concern which later is well documented (Fridrichsen,

Probleme 59ff) — Jesus must not appear as a magician.

4. THE PREDOMINANTLY MAGICAL CHARACTER OF THE GOSPEL STORIES

Unfortunately for the would-be apologists, not only the minor traits of the Gospel stories, but also the essential content of most of them come from the world of magic.

Jesus appears in Mk. as one possessed by a holy spirit and thereby made the son of a god; we have already seen the same sort of figure in the magical papyri ( PGM of Tyana, whom the Christians, and ancient opinion generally, considered a magician

— though his followers, like Jesus', denied the charge (Philostratus, Vita Apollonii

1.4,6; IV. 18; VII. 38; cf. Dio Cassius LXXVII.18.4; Origen, Contra Celsum VI. 41, with Chadwick's n3). At the beginning of his career Jesus, like Apollonius, was driven by the spirit (Mk. 1.12 and parallels; V.Ap. 1. 18) into the wilderness, where he was approached by an evil spirit but repulsed it (Mk. 1.13; Lk. 4.2flf; the mixture of traditions which represent Jesus now as possessing spirits, again as himself possessed (Eliade, Shamanism 33ff, esp. 236; Mk. i.i2f; 3.21-30; Mt. 9.34;

10.25; Jn. 7.20; 8.48^52; 10.20). On returning to settled areas Jesus, like Apollonius, became a wandering preacher with an attendant circle of disciples (Mk. 1.21,36;

3.7-19; Lk. 6.2off; V.Ap. IV. iff, 37; V.43 and passim ), and the admirers of both later appealed to their religious teaching and its success as evidence that they were not magicians [V.Ap. IV. 19, etc.; Origen, Contra Celsum 1.38,68, where he overlooks

Apollonius). Both, however, were distinguished from ordinary preachers by their miraculous powers; many of the same miracles appear in the stories about both of them; and most of the miracles reported of Jesus are those which are commonly reported of magicians and for which recipes are given in the magical papyri.

Thus we find (and the following references give merely a few examples of each) :

The power to make anyone he wanted follow him: Mk. i.i6ff; 2.14 (cf. V.Ap.

IV. 20) ; PGM IV. 1 7 1 6ff ; VII.30off,62off (from The Diadem of Moses) ; XIII. 238

(from The Eighth Book of Moses ) ; XXXIIa; etc.

IV.

5 10, 535). Other stories say he was born of a god (Mt. i.i8ff) ; so was Apollonius

THE BACKGROUND

Exorcism: Mk. 1.23,34; 3.11,22; 5. iff, etc.; V.Ap. IV.10,20,25, etc.; Lucian, Philopseudes 16; Origen, Contra Celsum 1.68; PGM I.115; IV.2170; V.gGff;

VII. 43 iff, etc.; Tamborino, Daemonismo i8f.

Exorcism at a distance, remote control of spirits, and the power to order them about: Mk. 7-25ff; Lk. 7.1-10; V.Ap. III. 38; Lucian, Philopseudes 13; PGM

I. i8off; ν.ι65ίΓ.

Miraculous cures: Mk. i.2gff (fever), 34, 4off; 2. iff; 3. iff (withered hand); 3.10;

5.25ff (issue of blood), etc.; Lk. 7. 1 8ff (lame, blind, etc.); V.Ap. VII. 39 (lame, blind, a man with a withered hand); Lucian, Philopseudes n(|| Mk. 2.12, he took up his pallet and walked); Origen, Contra Celsum 1.68; PGM XVIIIb;

XXXIII (fever); XXII. iff (issue of blood) ; VII.19 1—2143580 ( πάσαν νόσον , cf. Mt. 4.23), 677, etc.; Heim, Incantamenta ; DMP verso, passim.

Stilling storms: Mk. 4«35ff ; V.Ap. IV. 13; PGM I.120; V.137 (|| Mk. 4.41) ; XXIX.

Raising the dead: Mk. 5.21-43, etc. ; V.Ap. IV.45 (II Lk. 7. 1 iff) ; PGM XIII.278ff.

Giving his disciples power over demons: Mk. 6.7; PGM I.42ff, I93ff; IV.475ff, 732-747>85°» etc· Miraculous provision of food: Mk. 6.35ff; 8. iff; V.Ap. IV. 25; Origen, Contra

Celsum 1.68; Lucian, Philopseudes 35; PGM I.103-115; XIII.998; J. Sanhedrin VI.9(23c) and parallels.

Walking on water: Mk. 6.48; Lucian, Philopseudes 13; PGM I.i2off; XXXIV.

Miraculous escapes (his body could not be seized) : Lk. 4.30; Jn. 7.30,44; 8.20,59

(C, koine , etc.); 10.39; V.Ap. VII. 38; VIII. 30; PGM 1. 195-222; XXIIa.nf; XXXVI. 32off; Vienna National Library, Pap. gr. 323; Tamborino, Daemonismo

18.

Making himself invisible: Jn. 8.59; ΐ2·36(?); Lk. 24.31; V.Ap. VIII. 5; PGM I.i02,222ff,247ff; VIL6i9ff; XIII. 2686

Possession of the keys of the kingdom: Mt. 16.19; PGM I II. 541 ; IV. 1896

Foreknowledge:

Of his own fate: Mk. 8.3iff, etc.; V.Ap. XIII. 38, 41 ; PGM I\i73f; XIII.7ooff.

Of disasters coming on cities: Lk. io.i3f; 13.34^ 23-28ff; Mk. 13.2; V.Ap.

IV. 4; V. 13; Sibylline Oracles , passim.

General: Mk. 5.39; Jn. 1 1 . 1 iff; Mk. I4.i3ff; V.Ap. IV.24; V.24; VI. 32; PGM I.i88ff (|| Mk. 5.39); IV.231,250; V.288ff, etc.

Knowledge of others' thoughts: Mk. 2.8; 12.15; V.Ap. IV. 25; VI. 3; PGM I.176; IH-330,459; V.228; etc.

Metamorphosis: Mk. 9.3; PGM 1. 1 17^177; XIII.27off. The tradition that Apol lonius was the son of Proteus — V.Ap. I.4 — suggests that he was credited with this accomplishment, for which Proteus was notorious — Odyssey IV. 45517. How ever, in these metamorphoses the new forms assumed are disguises, by which the magician attempts to prevent recognition of his true and familiar form; in the transfiguration the new form reveals Jesus' true, supernatural powers, of which his familiar form was a disguise; cf. Odyssey XIII.22 iff,287ff. This

motif of the deity in disguise was common in Greek mythology: for example, the Homeric Hymn to Demeter 91—280, whence it influenced both the legends

CLEMENT OF ALEXANDRIA

of hellenistic cults — cf. Plutarch, De Iside 14-16 (357) — and the claims and practices of magicians, Lucian, Alexander 40; PGM IV. 1859. A form particularly important for NT thought was the descent-in-disguise legend, on which see above, p. 201. This is the more remote background of the transfiguration story, but its immediate background is in magical practice and apocalyptic theory.

From these came the notion that the magician could ascend into the heaven (or the mountain) of the gods and assume their glory: Is. 14.13^ Ezekiel 28.138^

II Enoch A 22.8ff; Ascension of Isaiah 7.25; 9.30; etc. In PGM IV. 475-750 and in the Hekalot we have instructions for such an ascent, either alone or with a disciple (732ff) - A disciple in whom suggestion produced the hallucination of such an ascent would have seen his master clothed in the garment of glory and talking with the inhabitants of the heavens.

Revealing supernatural beings to his disciples: Mk. 9.4; PGM IV.88ff, 1 72,7328^

897-922; V.1-40; VII. 549; DMP II. iff; VII. 10; XIV. 24f; etc.

Prescribing reforms of temple practices: Mk. 11.156!; V.Ap. 1. 16; IV. 1,23,24; etc.

Introducing a new rite, a meal by which his followers are united with him by partaking of food magically identified with his flesh and blood: Mk. i4.22ff;

Jn. 6.56; I Cor. 10.16; n.24f; cf. V.Ap. III. 25, 32, 51 ; see above, section A.

The role of magicians in introducing new religious rites is emphasized by Burkert, ΓΟΗΣ 3gf, and exemplified by Lucian, Alexander 38. Cf. Diodorus

V.64.4; Irenaeus — Harvey, 1.7.16! = Stieren, 1. 13. iff.

Claiming to be united with his disciples, so that he is in them and they in him:

Jn. 6.56; 14.20; 15. 36!, 9 (union in love); etc.; PGM XXXIIa: "Adona(i), Abrasax, Pin(o)uti and Sabaoth, enflame the soul and heart of . . . Amonius . . . for Serapiacus . . . now, now, quickly, quickly . . . Forthwith mingle together the souls of both and make . . . Amonius . . . one with Serapiacus . . . every hour and every day and every night. Therefore Adonai, highest of gods, . . . set to it, Adonai ! "

Claiming to be a god or a son of a god, or united with some god or supernatural entity (notably in statements beginning "I am"): Mk. 14.62; cf. 13.6; (Mt.

26.63, d νιος του θεοϋ του ζώντος C*W ΔΦ 090 al. ff2; DMP XX. 33, "I am the son of the god who liveth"); V.Ap. III. 18; Jn. 10.36, υΙος του θεόν et/xt. PGM

IV. 535, εγώ et/xt d υιός. Jn. 14.6, εγώ et/xt η . . . αλήθεια. PGM V. 148, εγώ et/xt η αλήθεια (cf. DMP IX.14). Jn. 8.12, εγώ et/xt το φως τού κόσμον. PGM XII. 232, εγώ et/xt d " Ηλιος ό όεόειχώς φως. Jn. I 7.2 I (to the Father), συ . . . εν εμοι καγώ εν σοί.

PGM VIII. 50 (to Hermes), εγώ ... συ και συ εγώ.

This list by no means exhausts the material. There are many other traits in the Gospels' picture of Jesus — particularly, but by no means exclusively, in the Johannine picture — which are common in magical material : the magician is the only one who knows his god (Lk. 10.22; Jn. 7.29; 17.25; PGM 1. 1 866! || Jn. 5.37; XIII. 5806!,

841-888) and is known by his god (Lk. 10.22; Jn. 10.15; PGM VIII.49); those who see him see the invisible god, of whom he is the visible image (Jn. 1.18; PGM

THE BACKGROUND

XII. 229, 235) ; and so on. But these similarities in sayings (and, a fortiori, mere similarities of terminology, like the uses of κύριος and σώζειν remarked by Robinson,

Text 252f) are less important for our purpose than the fact that the stories of the Gospels are mostly stories about things a magician would do. They are not mostly stories about things the Messiah would do. (Who ever heard of the Messiah's being an exorcist — let alone being eaten?) The closest parallels in the OT material are to be found in the stories of the prophets, but here, too, the exorcistic and the sacra mental elements are completely lacking. The sacramental side, too, is wholly lacking from the reports about the rabbis. Jesus doubtless was, as was Apollonius, involved in arguments about the proper observance of religious laws, he did behave like a prophet and was thought to be one, and he probably came to think himself the Messiah (certainly other people thought him so and their opinion cost him his life) ; but neither his messianic nor his legal opinions, nor even his imitations of the prophets, account for the most important of the stories about him, which are stories of a man who did the things magicians claimed to do. This fact was recognized in antiquity even by the Christian apologists: Justin, First Apology 30; Tertullian, Apology XXI.17;

Origen, Contra Celsum 1 1. 49 if.

Moreover, after Jesus' death his followers continued to credit him with typically magical activities. He appeared to his followers after his death (Jn. 20, etc.; V.Ap.

VIII. 3 1 |j Acts 9-3ff). He could, at will, make himself unrecognized or invisible

(Lk. 24.16,31; see above). He came through locked doors (Jn. 20.19,26; PGM XII.

279; XIII.io64ff; XXXVI. 3i2ff). He gave his followers power to handle serpents and drink poison without being harmed (Mk. 16. 18; PGM I.i 15; IV. 21 75 ; XIII. 253).

He gave them the holy spirit by breathing it into them (Jn. 20.22; Tamborino, Daemonismo 81, 102; Origen, Contra Celsum 1.68; PGM IV.3007-3085). And finally, he ascended into the heavens (Lk. 24.51 ; Acts 1 .pf ; V.Ap. III. 51 ; VI. 1 1 end; VIII. 30 end; PGM 1. 178; IV.475ff) and was worshiped as a god {V.Ap. VII. 21 ; VIII.5,7.vii,

3 1 end ; PGM 1 . 1 9 1 f ) and as a magician : Many sects in the gnostic wing of Christianity not only practiced magic, but remembered and revered Jesus as the great magician

(Anz, Frage 5-9; Irenaeus, Harvey, 1. 19. 2 — 1. 20 = Stieren, 1. 24-25; etc.). The apo cryphal gospels and acts, it is well known, add many magical traits to the picture of Jesus: e.g., Gospel of Peter 19; Gospel of Thomas (Leipoldt), heading, 13, 19, etc.; Acta loannis 87-105; Pistis Sophia , passim ; etc. Whether these traits came from tradition

(as is possible) or from invention (as is commonly taken for granted) or from both

(as is likely) they at all events show that many, perhaps most, Christians thought of Jesus as doing the things a magician would do in the ways a magician would do them. Even in the comparatively "orthodox " catacombs and mosaics he is customarily represented with the magician's rod (Goodenough IX.i6of, with recognition ol the Mosaic parallel but not of the reason for it).

5. THE RELATION OF "MAGICIAN" TO θβίος a VTjp AND " SON OF GOD5'

It is not to be thought that the destined occupants of the catacombs, or even most of the gnostics, would have described Jesus as a "magician. For his believers, he

CLEMENT OF ALEXANDRIA

was "the Son of God"; sceptical but reverent pagans would probably have called him "a divine man," θείος άνήρ. Both terms were applied to a good many wonder working religious teachers, some of them men who lived in the Roman empire, others ancient worthies like Pythagoras and Moses, whose stories were retold with the embellishments then fashionable. (For "Son of God," Wetter, Sohn, is still most valuable; the apologetic quibbles of Bieneck, Sohn, need not be read; Kramer,

Christos , non vidi. For the θείος άνήρ see Bieler, with the remarks of Nilsson, II .512 7fF ;

Levy ( Legende ) demonstrated the influence of the type on Jewish thought, but pushed his arguments too far; a recent recognition of its influence on the Gospels' picture of Jesus will be found in Robinson's Problem . . . Reconsidered 136F) In spite of the original diversity of these heroes and the diversity of theological explanations and consequent titles imposed on them, their stories, as now retold, follow a relatively uniform pattern (compare the "biographies" summarized in Heroes and Gods), but are not directly or even — in many cases — indirectly derived from each other. There fore the uniformity may in part be explained by the fact that the authors were describ ing a single social type, and Wetter's chapter "The Son of God as Antitype of the Magician" ( Sohn 73ff) leaves no doubt as to what that type was. The magician claims the powers which the "divine man" or "Son of God" is thought to possess.

The difference between these figures is one of social status and success. A wandering quack to whom servant girls go for potions or poisons has little chance to pass himself off (even to himself) as a deity in disguise. But let him begin to succeed and Lucian,

Alexander , will tell you how he can become the founder of a new cult and the spiritual director of Roman senators. He can then represent himself not only as the prophet, but also as the son, of a god ( Alexander 1 1, 39). He can also come to believe his own representations (especially if his success has not been achieved by such conscious, mechanical devices as Alexander used, but has resulted from abnormal elements in his own personality which he himself would understand as evidence of supernatural powers). His enemies, however, will continue to call him a magician (γόης, Lucian,

Alexander 1, cf. Burkert, ΓΟΗΣ ; μάγος, Alexander 6, cf. Nock, Magus', the different connotations of the words do not here concern us. For alternation of the terms προφήτης

(by friends) and γόης (by enemies) see Nock, Alexander 162.) "Magician" thus covers a larger range than θείος άνήρ or "son of a god"; it includes the market quack (Origen, Contra Celsum 1.68) and even the individual who practices in private (Lucian, Lucius 4), whereas the θείος άνήρ or "son of a god" must be a man of considerable public reputation and theosophic pretensions. With the difference of pretensions goes a supposed difference of technique which we have already noticed: the hole-in-the-corner magicians peddle spells and ceremonies and materia magica, therefore they have a different set for each different purpose (and customer). The θείος άνήρ has his own spirit who has only to be ordered and will at once obey. The son of a god is himself a spirit who can command others. To guess what these differences meant in terms of actual experience is a risky, albeit tempting, game. One may begin with the notion that "miracles" (including magical cures and the like), when they do happen, are usually the results of suggestion (often hypnotic) and therefore depend both on the practitioner's "power of suggestion"

THE BACKGROUND

and the patient's "suggestibility" (for which the ancient terms — of equally obscure meaning — were "spirit" and "faith"). Both of these can be augmented by various factors among which are rituals, practice, previous success, and consequent reputation.

As the two latter increase, the dependence on rituals diminishes and the practitioner also rises in social position, self-esteem, and consequent claims. The change is a gradual one and the magical papyri therefore contain a medley of material, from simple charms for particular purposes through the more complex rites necessary to gain control over one or another spirit, to the elaborate initiations which will enable the magician himself to enter the heavens, become a god or the son of a god, and command all the spirits inferior to himself. The difference reflects that between the do-it-yourself world of the peasants and the slave service available to the rich.

Theologically (or, demonologically ?), however, as the magical texts show, this difference is one of form, not of essential content. And this fact is reflected by the terminology. Once the requirements of social status and decorum are met, the same man will customarily be called a θβΐος άνηρ, or son of a god, by his admirers, a magician by his enemies. Within this area all three terms refer to a single social type, and that type is the one characterized by the actions listed above, which make up by far the greatest part of the Gospels' reports about Jesus.

6. JEWISH AND PAGAN OPINIONS OF JESUS

From what has been said it follows that Jesus would be described, by those who did not accept his claims, as a magician. This description in fact appears repeatedly in both the pagan and the Jewish traditions about him, which have generally been disregarded by Christian scholarship on the grounds that they were pagan or Jewish.

Yet surely the historical study of any man should take account of, not only the reports of his adherents, but also those of his opponents, and especially those of comparatively indifferent observers.

The judgment of Jesus' opponents is reported already by the Gospels: Mk. 3.22 and parallels. The scribes who came down from Jerusalem said, "He has {control over) Beelzebub," and "he drives out demons by the ruler of the demons." A different version of the same tradition appears in Mt. 9.32. From Mt. 10.25 T seems that Jesus was actually nicknamed "Beelzebub." Other traces of the same tradition appear in the arguments to refute it, which Mt. 12.27 and Lk. 11.19 derived from ζΤ Other, independent, traditions to the same effect, but ignorant of Beelzebub, are stated or reflected in Mk. 3-28f; Lk. 12.10 and parallel (probably QJ ; possibly

Mk. 6.14 (see Kraeling, Necromancy ); and certainly Jn. 7.20; 8.48,52; 10.20 (on the typical alternation between possession and being possessed see above, section 4, and also Samain, Magie 475). It is significant that neither Jesus nor his followers denied the charge that he "had" a spirit; on the contrary, they admitted it, but claimed that the spirit was a holy one (Jn. i.32ff; Mk. 1.10 and parallels; 3-28f and parallels; etc.; and The Gospel according to the Hebrews ). This charge would explain

Jesus' "recurrent warnings against being 'scandalized' at him, or being ashamed of him, or denying him before men" which appear in both Jn. and the synoptics

CLEMENT OF ALEXANDRIA

and were thought perhaps historical by Dodd ( Historical Tradition 221), though he was unable to suggest an adequate reason for them. After Jesus' death the charge that he had been a magician continued common in both Jewish and pagan circles

(e.g., Justin, First Apology 30; Dialogue 69.7; Origen, Contra Celsum 1.6; Tertullian,

Apology XXI.17; Arnobius, Adversus Nationes 1. 43; B. Sanhedrin 43a as quoted by Strack, Jesus 1; the Talmudic references to Jesus as Balaam (Bileam), e.g., Strack,

Jesus 5; J. Shahbat XII.4(i3d) — ben Stada is probably Jesus — cf. Origen, Contra

Celsum 1.28; Krauss, Leben 40IT; the Clementine Recognitiones 1. 58). B. Sanhedrin

107b and B. Sotah 47a represent Jesus as a pupil of Rabbi Joshua ben Perahyah who appears in Babylonian magic as a great magician who had ascended into heaven and mastered all the demons, Montgomery, Incantation Texts 225fT and texts 8, 9,

17, 32, 33. Birds of a feather . . .

These accusations were of course denied but were not fundamentally countered by Jesus' followers, of whom some (as remarked above, section 4 end) not only admitted, but celebrated his achievements as a magician, while all perpetuated and many added to the magical traits of the stories and pictures in which he was represented.

Once again, the question was not his control of spirits, but only the means by which he had achieved it.

This evidence is further strengthened by that of comparatively indifferent observers

— both Jews and pagans — who did not become followers of Jesus, but admitted his magical powers and attempted to make use of them in their own operations. Even in his own lifetime, if we are to believe Mk. 9.38^ men who were not his followers were using his name to cast out demons. Similar usage continued in Palestine as late as the second century, when T. Hullin 1 1. 2 2 and parallels show us one Jacob, of the village of Sama in Galilee, offering to cure snakebite "with the name of Jesus." This does not prove Jacob a Christian! Jesus' reported magical powers led pagan magicians also to use his name in their charms: PGM XII. 192 and 391b

(For discussion and additional examples see Eitrem, Demonology 4-9.) Bickerman called to my attention Augustine, De consensu euangelistarum I.i iff, esp. iqf: there are pagans who think Jesus a man who attained the highest wisdom; they attribute to him letters to Peter and to Paul ( !) pretending to teach the magic ( exsecrabiles ) arts by which he did his miracles. Particularly important is the story in Acts 19.13 fT, that Jewish exorcists in Ephesus were using the name of Jesus in their exorcisms by a.d. 55 and that some of them used the formula "I conjure you by the Jesus whom

Paul proclaims." To this the demon is represented as replying, "Jesus I know and Paul I am acquainted with, but who are you?" This is intended as testimony that Jesus and Paul are supernatural powers prominent in the demonic world: Jesus because he has the holy spirit, and Paul because he has Jesus. That the demoniac thereupon jumped on the would-be exorcists and drove them away in disgrace is said to have produced great reverence for the name of Jesus among "all the Jews and Greeks dwelling in Ephesus . . . And many of the believers <(the Christians) began to confess and declare their magical practices. And a large number of those who had been practicing the occult arts brought together their books and burned them publicly, and when their prices were reckoned up they came to 50,000 pieces

THE BACKGROUND

of silver." Here the writer's intention is not only edification but also admonition.

What had happened in Ephesus was to be an example to the churches which the writer had in mind — presumably those of Asia Minor. He suspected or knew that they were thick with magicians, and — as Paul in Colossians — he reminded them that Jesus is more efficacious in magic than are the demonic powers. There is also a strong element of party propaganda: the only safe use of Jesus' name is that by proper representatives of the Pauline party.

7. MAGIC IN THE PRACTICE OF JESUS5 FOLLOWERS

It must again be insisted that the practices and teachings of a man's followers are at least some indication of what he practiced and taught. If so, the accounts of Jesus' followers indicate strongly that he practiced and taught magic. His followers are said to have begun exorcising demons and performing cures already during his lifetime, and of course by the use of his name (Mk. 6.7,13,14 and parallels; Lk.

10.17). After his death they reportedly continued and developed these practices (Acts 3.6; 5.5,9,15^ etc.). Paul is credited with them, too, and also— as were Jesus and Peter — with miraculously effective curses, another hallmark of the magician

(Acts 3.6; 5.5,9,15^ 1 3·9ίΓ; i4.8ff; i6.i6ff; Mk.i i.i3ff,2off and parallels; Lk. 9.54;

PGM XIII. 248, 26 iff; etc.) Paul wrote to the Corinthians that he had been present in Corinth in the spirit and so had judged and "handed over to Satan, for destruction of the flesh" — that is, for affliction with some illness, a common magical procedure

(see Audollent, passim) — a libertine member of the Corinthian church (I Cor. 5.3^6) .

Consequently, Alexander and Hymenaeus were "given over to Satan" by pseudoPaul in I Tim. 1.20. Here, probably, is to be found the explanation of Paul's claim to have some strange power over his congregations (I Cor. 4-igf; 5.3^6 ; II Cor. 6.7; io.3ff; 1 2. 1 9-1 3.7), as well as his fear that the initiates in Jerusalem might invalidate his work (Gal. 2.2 and my comments in Problems 1 i6f). The keys given to Peter could lock as well as open (Mt. 16.19). Also, like Jesus, both Peter and Paul were believed to have such supernatural powers resident in their bodies that even the shadow of Peter (Acts 5.15) or clothes which had touched the body of Paul (Acts 19.12) might work wonders. The great importance of exorcism in Jewish Christianity is noted by Schoeps, AF% 65b The widespread practice of magic in the gnostic wing of Christianity is notorious from the accusations made by other Christians (Irenaeus,

I passim ; Hippolytus, Philosophumena , passim) and Bauer, Rechtglaubigkeit , has given strong reasons for his opinion that in the late first and early second centuries the gnostic wing was larger than that which subsequently became "orthodox."

Far more important, however, than such occasional phenomena as exorcisms, blessings, curses, and cures, more important, also, than the extraordinary gnostic ceremonies of which the self-styled "catholics" complained, was the essentially magical nature of the fundamental rites of initiation (baptism) and communion

(eucharist) by which practically all Christian communities were constituted and held together. To the Christians, of course, these rites were the sacred mysteries instituted by the Son of their God, but this is no argument against their magical

CLEMENT OF ALEXANDRIA

character. The parallel rites are also represented by the magical papyri as sacred mysteries, divinely revealed. The magical character of baptism is made particularly clear by Paul's reference in I Cor. 15.29 to "those who are baptized on behalf of the dead" — a piece of substitutionary magic Paul evidently approved, since he used its unquestioned efficacy as an argument to prove that the dead must be raised.

This sort of thought about Christian ceremonies continued down to the time of Clement, who objects to women wearing wigs because, when the presbyter lays his hands on a wig and gives his blessing, the blessing will not go to the wearer of the wig but to the head which provided the hair (1. 27 1.20 — this was pointed out to me by H.C.).

The fundamental position of baptism and the eucharist testifies that this sort of thought was not secondary in Christianity, but primitive. Such, too, is the testimony given by the essential character of the stories in the Gospels, reviewed above. The same conclusion follows from Paul's account of Christianity as, essentially, salvation by possession; this, too, has been pointed out above, but we may add here the evidence afforded by the practice of "speaking with tongues" in the Pauline churches. As I Cor. 12 makes clear, this was the utterance of incomprehensible sounds, thought by the believers to be the speech of the spirit which possessed the speakers. As already remarked in section IX this is a common symptom of schizophrenia. In the church at Corinth it had become so common that it was disrupting the worship, and those especially gifted with it were giving themselves insufferable airs. Therefore Paul has to insist that it is not the greatest of the gifts of the spirit. The spirit gives also wisdom, knowledge, faith, and the ability to perform cures, do miracles, prophesy, distinguish good from evil spirits, and interpret things spoken in "tongues." And practically all of these Paul treats — on this occasion — as superior to the gift of "tongues." This obviously does not reduce the importance of magic in Paul's view of the world. On the contrary, it extends it to the whole of normal, as well as abnormal psychology. Everything good is the work of "the spirit." Another thing to be noticed is that even in this letter Paul thanks God that he speaks with tongues "more than all of you" (14.18), and elsewhere, when this problem of church discipline is not uppermost in his thought, he treats "speaking with tongues" as the supreme prayer of the Church — Rom. 8.26f. τδ γάρ τί προσευξώμεθα καθο άεΐ ούκ οίάαμεν, άλλα αυτά το πνεύμα ύπερεντυγχάνει στεναγμούς άλαλητοις" ό δε ερευνών τάς καρά ίας οίάεν τί το φρόνημα του πνεύματος , ότι κατά θεάν εντυγχάνει ύπερ αγίων, αλάλητος is not "unutterable" — they were uttered — but "inarticulate"; vs. LSJ see Bauer, Wb. ad loc. This fact, and the phenomenon of glossolalia, were tactfully overlooked by Dietzel, Beten , whose denial of the relation between Christian and magical prayer overlooks the main problem — to explain, not why the spirit spoke to the churches, but why it spoke to them in jabberwocky. στεναγμός was noted above (section 3) as a characteristic form of magical utterance. Such στεναγμοί αλάλητοι are preserved in great quantities in the magical papyri, of which they are perhaps the most notice able peculiarity. These are the "magical words," of which a few are fixed formulas with secret significance, but most are apparently meaningless combinations of letters by which the magician calls the spirit in the spirit's own language, or, by speaking

THE BACKGROUND

in the spirit's language, shows that the spirit is already in him and acting through him. These combinations perhaps were not intended merely for repetition, but were indications of the "tune" — the sort of phonetic combinations and interspersed names — appropriate to the spirit invoked.9 A number of examples of this have been given in the passage already quoted from PGM in section i ; more are to be found on almost every page of that collection. The same sort of thing appears in Pistis Sophia in the concluding prayers of Jesus, presumably representative of the secret prayers of the church that preserved this material: "Then (after the resurrection) Jesus stood with his disciples . . . and called on (the Father with) this prayer . . . 'Hear me, my Father, Thou Father of all paternity, Thou endless light, αεηιουω' ϊαω· αωϊ' ωϊα'10 φινωθερ' θερνωφ' νωψίτβρ'11 ζαγουρη' παγουρη'12 νβθμομαωθ' νεφιομαωθ' μαραχαχθα ' θωβαρραβαυ' θαρν αχαχαν' ζοροκοθορα' Ϊ€ου'13 σα/3αω#.'"14 (ch. 13b? my translation from Schmidt-Till; more in ch. 142.) Harnack, Pistis-Sophia 89, recog nized these passages as representations of speaking with tongues; Anz, Frage 8, as incantations, combinations of "magical words" typical of the spells in the magical papyri. Both were right. The spirit which spoke through the Christians and the spirits which spoke through the pagan magicians spoke the same characteristic language.

Not only is this fact recognizable now; it was already recognized in antiquity and its recognition resulted in many of the persecutions which Christianity every9. Similar variation of formulas appears in the defixiones and on the magical gems. In the case of the latter, this has been noticed by A. Delatte and P. Derchain, Les Intailles magiques greco-egyptiennes, Paris

(Bibliotheque Nationale), 1964, p. 234: "There is little chance, naturally, that one should ever find an object <magical gem) exactly in agreement with the ancient description <in the magical papyri), for magic was still living at the time of the redaction of the papyri and at the time when the gems were engraved. Consequently, on comparing them with each other, one finds that the formulas show in numerable variations about which the magicians hardly bothered at all."

10. Such variations on the seven vowels are common in PGM, in which all t^iese individual com binations occur — e.g., no. II. lines 14-16; III. 230, 436; IV.487,917, etc.; XII. 72, 102,1 19; XIII. 209,

895,936; and often elsewhere. Ίαω (the commonest Greek name of ΠΊΓΡ) is particularly frequent. In the indices to PGM, which the late Prof. Preisendanz kindly put at my disposal , there are three col umns full of references to 7αω — far more than to any other deity. (The runners up are Άδωναϊ and Έρμης with about two columns each.) 1 1. PGM III.l86a-b νωφιθ^ρ] θερνωφι; IV.828f φινωθερ νωφιθβρ θζρνωφι.; VII. 316 φινωθ ββρ.

12. PGM V.479; XXXVI. 308; 348-353; Kropp, ΑΚΖ no. XIII line η, cf. vol. III. 127 n5. All these give the combination ζαγουρη παγουρη or vice versa (with various misspellings). Both words also occur alone; e.g., PGM XXXVI. 10,64 ζαγουρη; VII. 597, 606 -παγουρη . ΐ3· PGM XI 1 1.959-963 νβθμομαω μαρχαχθα . . . θαρνμαχαχ ζαροκοθαρα . . . η€ου θωβαρραβαυ

Audollent no. 267. 9^ [ ν]€θμομαω μαρχαχον . . . ζαρακ[α]θαρα θωβαρραβαυ θαρναχαχα. νζθμομαωθ (and variants) alone: PGM II.119 (μαϊραχαχθα follows in 120); HI. 153 (οροκοτοθρο in 154); Audollent no. 242.31. vet/ncombinations do not appear elsewhere; this is probably a corruption. θωβαρραβαυ (and variants) alone: PGM VII.977 ; XIXa.42; Audollent no. 242.18; etc.

i€ov alone: PGM II. 137; VII.476; XII. 189; XIII.810; etc.

14. PGM VII.605; IX. 7; XII. 207 and passim (almost a full column of references in Preisendanz' index).

CLEMENT OF ALEXANDRIA

where called forth. These persecutions require explanation both because of their frequency and because of the general tolerance throughout the Roman empire for cults of oriental gods and deified men. Occasional exceptions to this tolerance might be explained by peculiar local circumstances; but the consistent opposition to Chris tianity evidently resulted from something characteristic of the new religion. What was it? The common answer is, the Christians' refusal to worship other gods. But other worshipers of Yahweh — the Jews and the Samaritans — also refused to worship other gods, and they were not generally persecuted. Consequently, the Christians had to explain the persecutions as inspired either by the demons or by the Jews who, they said, denounced them to the authorities (Acts 13.50; 14.2; 17.6,13;

18.12; etc.; I Thess. 2.15; Apoc. 2.9; Martyrdom of Polycarp XII. 2; XVII. 2 ; Justin,

Dialogue 17). But for what could the Jews denounce them? Certainly not for refusing to worship other gods — that was the hallmark of Jewish faith generally. Perhaps for political conspiracy? But there is no evidence that the Christians were, generally, subversive in politics — not, at least, after the year 70 (cf. Brandon, Fall) — and they were not generally accused of plotting political revolution. What they were accused of was the practice of magic and other crimes associated with magic : human sacrifice, cannibalism, and incest ( Vita Apollonii VIII. 5; Bidez-Cumont, Mages 78ff). The accusation of magic probably appears already in Suetonius, Nero 16.3: Affiicti suppliciis Christiani , genus hominum superstitionis novae ac maleficae , since a maleficus is par excellence a magician. Suetonius indicates how Tacitus, Annals XV. 44, should be interpreted : quos per flagitia invisos vulgus Christianos appellabat . . . exitiabilis super stitio . . . atrocia aut pudenda . . . primum correpti qui fatebantur. What did they — under torture — confess? Probably firing the city, probably being Christians, but probably also magic. Magic figures conspicuously in charges against Christians from the second century on (Origen, Contra Celsum 1.6; VI.4of; Passio Perpetuae 16; Acta

Aeacii V.2f; Tertullian, Ad uxorem II. 5. 2; Eus., HE III. 26. 4; IV.7.iof). Moreover, as the passages from Eusebius show — and they could be paralleled by many more from Irenaeus, Hippolytus, and Epiphanius — the Christians made considerable use of this charge against each other. Presumably they knew what they were talking about. Magic has been commonly neglected in discussions of the persecutions

(recently Speyer, Vorwurfen\ Gregoire, Persecutions', Sherwin-White, Letters , 69 iff,

772ff; but cf. Freudenberger, Verhalten , i65ff) because these have commonly been based on the apologists and the acts of the martyrs and have neglected the heresiologists. But the apologists chose to defend Christianity from charges of which most

Christians were not guilty; magic would have been an embarrassing topic, therefore they do not mention it and in refuting the charge of cannibalism they are careful to avoid the question of what actually happens in the eucharist. And the acts of the martyrs are propaganda pieces, mostly intended to represent the Christians as innocent victims martyred solely because of monotheism; therefore they usually say nothing of any of the flagitia cohaerentia nomini (Pliny, Epistulae X.96.2). Commonly

the only questions are, are you a Christian, and, will you sacrifice to the gods or the emperor. References to magic are rare, but references to incest and cannibalism are even rarer. History should not use such material uncritically.

THE BACKGROUND

8. RECAPITULATION AND CONCLUSIONS

It has now been argued: (i) The fact that Jesus is not represented as a magician by the Gospels is insignificant; "magician" was a dirty word. The significant fact is that he is represented as the possessor of the holy spirit and as "the Son of God," a supernatural being recognized by demons as able to command them; he is repre sented as a successful magician would have represented himself. (2) That Jesus is not represented by the Gospels as using long spells or magic rites is insignificant.

Once a magician "had" his spirit, he need only command and it would instantly obey. Here too, the Gospels represent Jesus as a successful magician would have represented himself. (3) The miracle stories are shot through with minor traits of magical practice. (4) The Gospels' stories generally are stories of things magicians claimed to do, and they add up to an account of a magicians' life: Jesus' career began when he was possessed by a spirit which drove him into the wilderness (Mk. i. 9-1 2; Eliade, Shamanism 33f, 64, etc.). After surviving the ordeals to which the spirits there subjected him (Mk. 1 . 1 3f; Lk. 4.1-13 and parallel) he returned to Galilee, where he made his reputation as an exorcist (Mk. 1.2 1-27) and miracle worker (Jn. 2.1 1) and developed it by cures of which the magical traits have been mentioned. He then empowered his disciples to exorcise and perform similar cures

(Mk. 6.7-13). His fame thus became so great that magicians outside the circle of his followers began to use his name in their exorcisms (Mk. 9.38^. He introduced a reAerJ — a secret meal in which his followers were united with him by being given bread and wine which were declared to be his flesh and blood (Mk. 14.17-25).

In all these respects his work can be paralleled from the claims and careers of other magicians. (The career most fully reported is that of Apollonius of Tyana, but if the facts were known a closer parallel would perhaps be found in that of Jesus' contemporary Simon, the Samaritan magician, who also did miracles, claimed to be a power come down from heaven, and was credited with the introduction of mystery rites; Acts 8. gif; Irenaeus, Harvey, 1. 16. iff = Stieren, 1. 23. iff; see my Account ).

Finally, Jesus claimed to be the Messiah and, somehow, to be able to admit his followers to the kingdom of God, and this claim resulted in his execution by the Romans. In these respects, too, his career has magical parallels. Josephus reports that in Palestine at this period there was a plague of messianic magicians who similarly raised men's hopes for the coming of the kingdom, and who came to similarly bad ends at the hands of the Romans: BJ II.258ff; AJ XX. 97, 188; Eisler,

Messiah 328,360. (5) and (6) As in the case of Simon and Apollonius and other magicians, Jesus' followers believed him a supernatural being — either a "divine man" or the son of a god — but by his enemies he was repeatedly accused of magic, and these accusations were continued in later reports. Magicians also believed him a magician. (7) Similar accusations were often made against his followers and were an important contributory element in a long series of legal actions against them.

Considerable groups of his followers admittedly did practice magic and claimed that he was the source of their practice.

In addition to all this evidence, the hypothesis that Jesus practiced magic helps

CLEMENT OF ALEXANDRIA

to explain some of the major problems raised by the Gospels' account of his career.

From that account it appears, as we have said, that Jesus claimed to be the Messiah and bringer of the kingdom of God. But there is no clear indication of his functional relation to the kingdom as already present. This is the more surprising because even in the sketchy stories given by Josephus (in the passages just cited) about the other messianic magicians, their functions in relation to the kingdom are clear — they will use their magical powers to protect their followers and overthrow the Romans. So, too, the functional relation of the Baptist to the kingdom he foretold was clear (above, pp. 205ff). So, the silence of the Gospels on Jesus' functional relation to the kingdom is amazing. It requires an explanation.

A considerable number of Jesus' sayings imply that not only he, but also some of his disciples are already in the kingdom (above, pages 21 iff). A few of these sayings indicate that there is a definite, practical way to get in. That the mystery of the kingdom has been given to the disciples (Mk. 4.1 1) suggests there was some initiatory rite. Some evidence indicates that this rite was a baptism (above, pp. 1 78fT) , which Jesus administered secretly to a chosen few of his followers (above, pages

209IT). Notice how the Q statement, Lk. 7.28 and parallel, fits together with Jn.

3·3ίΓ. ζ) says, none born of women is greater than the Baptist, but the least in the kingdom is greater than he. We naturally ask: What, then, was the least in the kingdom born of? John answers, of water and the spirit, for only those born of water and the spirit (as opposed to those born of the flesh) can enter the kingdom.

"Of water and the spirit" is evidently a reference to Jesus' baptism, which gave the spirit, as opposed to the Baptist's baptism, which did not (above, pp. 207, 219).

But the gift of the spirit is not prima facie identical with admission to the kingdom, so we ask: How did Jesus' baptism admit his followers to the kingdom? If they

"entered" the kingdom, the kingdom was presumably an area in space. This agrees with what we have seen, that the kingdom was par excellence in the heavens (above, pp. 202ff). And we have also seen some reason to believe that there was in Palestine in Jesus' time a magical technique for ascending and causing others to ascend into the heavens (p. 181). If we suppose Jesus practiced such a technique, then we can explain the famous ζ) statement that, "the Law and the prophets were <fn force) until John (the Baptist); since then the good news about the kingdom of God has been proclaimed and anyone can force his way into it" (Lk. 16.16; Mt.

1 1. 12; notice the connection in Mt. with the statement that the least in the kingdom is greater than John). Here Dodd's despairing comment on Mt. 1 1.12 ("the kingdom of heaven suffers violence"), "whatever that may mean," (

Parables 40) and such fantastic explanations as Danker's {Lk. 16.16) are alike unnecessary. In the magical techniques of ascent to the heavens the magician must overcome the resistance of the demonic or angelic guards who bar the way (e.g., Hekalot , passim). These powers

Jesus has overcome (Col. 2.9ff). Therefore the kingdom of God, the kingdom in the heavens, suffers violence, because its guards are now overcome by those who have been magically united with the spirit of Jesus in the baptism of Jesus — "the mystery of the kingdom of God," the mystery which enables them to force their way into the kingdom of God in the heavens.

THE BACKGROUND

It was while performing such a baptism that Jesus was arrested. The rite was secret. He chose a lonely garden, through which a stream still flows, and went there late at night, after the ceremony of the eucharist had assured the magical union of his circle of initiates. Since he did not wish to be interrupted (this is essential in magic) he set guards (Mk. 14.32-34). He had no intention of being arrested if he could help it. The agony, therefore, has no likelihood, and it was witnessed by no one. (On its homiletic motivation see my Comments 22f.) When the guards fell asleep and the police arrived unexpectedly they surprised both Jesus and the initiate, νεανίσκος τις . . . περιβεβλη μένος σινδόνα επί γυμνόν (Mk. 14.5 0 — the proper magical costume in the proper magical setting. If this was not an initiation, what was the young man doing with Jesus at such an hour, in such a place, and in such a costume ?

The author of Acta Ioannis 94ff understood what was going on and supplied an account of the initiation, unfortunately imaginative.

D. The rite was a means of ascent to the heavens 15 By this time the reader will have forgotten that the long argument about magic was only the third point of a yet longer argument to show that important elements in the Pauline concept of baptism were derived from Jesus. That argument went as follows: Paul interpreted baptism as a means of uniting with Jesus — the one baptized was possessed by Jesus' spirit. This notion looks as if it came from Palestinian demono logy. And there are five traits which point to Jesus as its source. Of these five we have now considered three: (A) as a means of uniting with Jesus baptism is essentially like communion, which he introduced; (B) it is effected by the spirit, and the spirit is distinctive of Jesus' ministry, as opposed to that of the Baptist; (G) it is a magical ceremony, and Jesus practiced magic. These points established, we can now go on to the fourth : (D) : Pauline baptism was conceived as a means of ascent to the heavens (above, pp. 2i4ff) and in Christianity this notion of ascent seems also to go back to Jesus.

Here I shall try to show: (1) the notion of ascent to the heavens was an important element in Jesus' Palestinian background and had led to the development of a tech nique for ascent which Jesus might have practiced. (2) There is a good deal of indirect evidence indicating that Jesus did practice such a technique. (3) A number of passages in the Gospels are best understood as records of his practice and consequent teaching.

15. The following section was written before Scholem brought to my attention Danielou's Traditions , which has demonstrated the existence in Christianity, from Paul's time on, of a secret tradition concerning ascent through the celestial spheres. Danielou supposes the tradition came into Christianity through Paul, from his Jewish background (p.213). But it is too widely distributed to have come through Paul alone. Most important is the extensive agreement of the conclusions of two studies made in

dependently and from standpoints so diverse as Danielou's and my own. For this reason I have not here repeated Danielou's arguments and evidence, but leave it to the reader to look at his article and see for himself at once the independence of the methods and the agreement of the results.

CLEMENT OF ALEXANDRIA

I. BACKGROUND

The notion of ascent to the heavens is ancient (Kees, Himmelsreise; ANET ioif; 1 1 8 ; 446f; Oppenheim, Interpretation 259, 267, 282, 287) and widespread (Eliade,

Shamanism 1 15-144, 181-288, 392ff, 477IE, 48788, etc.). Since 1897, when its funda mental importance in gnosticism was pointed out by Anz, Frage, it has been repeatedly discussed; see the remarks of Lewy, Oracles 41388 The studies cited above (section

IX, end), and those referred to below are only the ones I have found most useful for the present argument. I have been careful not to refer to some others, particularly the pan-Iranian anachronisms of Reitzenstein and the inaccurate work of Widengren.

Many of the previous discussions have dealt with origins and transmission. Here we need concern ourselves only with those forms in which the notion of ascent to the heavens may have been current in the Greco-Roman, Hebrew, and Aramaic material circulating in Palestine during the first centuries b.c. and a.d.

In the eastern Mediterranean world the notion had early appeared in two forms not always separate, but roughly distinguishable. One was that of the soul's ascent into the heavens after death, the other that of ascent by living individuals either carried aloft bodily or in dream or ecstasy leaving their bodies below and returning to them later (cf. Bousset, Himmelsreise 136). It would be a mistake to try to differen tiate the stories too sharply. Paul, who claimed to have had such an experience, said he did not know whether, at the time, he was in the body or not (II Cor. 12.2).

The notion of the soul's ascent at some time after death was given classical expres sion by Plato and was popularized by Posidonius and by the influence of astrology

(Gumont, Lux ch. 3). During the first centuries b.c. and a.d. it was widely held all over the Greco-Roman world, including Palestine where the expectation that the righteous dead will shine as the stars appears in Dan. 12.3 and their ascent to heaven is anticipated in many other apocalypses (e.g., Enoch 39.4^ 104.2; Assumption of Moses io.8ff; II Enoch 8-9; II Baruch 51.10IT; III Baruch 10.5; IV Ezra 7.90-100). Josephus attributes the same anticipation to the Essenes ( BJ II. 1 5488) . It thus provided a widespread background for the other notion — more important for our present purpose — of ascent to the heavens by living men.

Claims of ascent by living men seem to be attacked already in Is. 14.13k and Ezek. 28, esp. verses 1,6,9,13-17. With "the mountain of God" cf. Mt. Sinai in the story of Moses' enthronement as κοσμοκράτωρ (Ezekiel tragicus in Eusebius,

Praeparatio Euangelica IX.29.4f) and the mountains of the temptation and the trans figuration (Mt. 4.8; Mk. 9.2-13; and parallels) and notice how many early Christian revelations were located on mountains — Apocryphon of John , Pistis Sophia, Gospel of Eve , Great Questions of Mary (Epiphanius, Panarion XXVI. 3.1; 8. iff) etc. The OT attacks show that those who ascended were thought to become like the gods in form

(cf. Phil. 2.6) and to be enthroned in the heavens (cf. Col. 3.1; Heb. 1.3; this is the goal of the Hekalot mystics). A similar claim is made on behalf of (probably)

Simon Maccabaeus in Ps. 110.1: "The Lord said unto my Lord, 'Sit thou at my right hand.'" (This is perhaps related to the stories of the enthronement of Moses, cited above, and the ascension of Levi, Testaments of the Twelve Patriarchs, Levi II. 6—

V.3.) Such notions were so important to the religious and political thought of the

THE BACKGROUND

times that they were made the basis for the imperial cult. Indeed, from the beginning of the second century a.d. the Roman government regularly used magical techniques to effect the bodily apotheosis of the dead emperor (Bickerman, Kaiser apotheose esp.

5ff). Stories like those of the bodily assumptions in supernatural chariots of the wonder-worker Elijah (II Kgs. 2.1 1) and the magician Medea (Roscher, Lexikon s.v., 11.2.2484!!) had been allegorized already by Parmenides as types of the ascent in philosophic contemplation (Diels, 28 Bi). This allegory was powerfully developed by Plato in the Phaedrus 244-257, and the type became a commonplace (Pippidi,

Recherches 170 n2). For the great importance of the Elijah story as an archetype of the Christian experience, see Schrade, Ikonographie 81-89, where the connections with magic are made clear. Besides these myths specifying the means of transportation, there were simpler stories deriving directly from ecstatic or cataleptic experience; and these, too, Plato made part of the literary tradition known to all educated men by his use of the story of Er in the Republic (6 1 4ff ) as the vehicle for his great myth of the travels of the soul (cf. Dieteridi, Mithrasliturgie 199). Similar pictures of the heavens and the afterlife, as seen by living travelers, abound both in subsequent Greek and Latin literature (for example, from the first century b.c., Cicero's Somnium Scipionis ), and appear in Judaism both in the work of Philo (e.g., De opificio 23 end) and in the originally Hebrew or Aramaic Palestinian apocalypses from the late third or early second century b.c. and thereafter ( Vita Adae 25; Enoch 14.8; 39.3; 71.1,5;

II Enoch 3.1 ; 67.1 ; Testament of Abraham 10; Apocalypse of Abraham 15; Testaments of the Twelve Patriarchs , Levi II.6ff; III Baruch 2ff; etc.). For the connection of these with Christian and rabbinic traditions see Scholem, Gnosticism ch. III.

These Palestinian stories of heavenly travels probably reflect not only the influence of Greek literature and near eastern legend, but also, as has long been recognized, some sort of experience deliberately cultivated in pietistic circles (Eppel, Pietisme

66f; Scholem, Gnosticism ch. III). Just when this deliberate cultivation first took the form of a definite technique there is no telling. In the Qumran Hodayot and Manual of Discipline the entrance of the members of the sect into the company of the angels is often celebrated as something already accomplished. The members are "together with the angels of the presence, and there is no need of an interpreter between them" ( Hodayot VI. 13; cf. III.2off; Xl.ioff; XIII. 27ff; frag. 2.ioff; frag. 5; Manual of Discipline XI. 5-10). 16 It is not unlikely that this state of affairs was thought to be brought about by some special ceremony, probably that (άπ)αθανατίσμδς referred to by Josephus, AJ XVIII. 18: άθανα τΐζονσί τας φυχάς, περίμάχητον ηγούμενοί τοΰ δίκαιον την πρόσοδον. Bousset, Himmelsreise 143? remarks that this probably refers to demonic opposition, encountered by those who would enter the heavens, and explains the Essenes' desire to know the names of the angels, which commonly serve as passwords for the journey ( Hekalot , passim ; Josephus, BJ II. 142; contrast the forced translations and implausible conjectures discussed by Feldman, Josephus , at AJ XVIII. 18). περίμάχητον here has the same background as Eph. 6.1 iff, "Put

16. i Q_28b, cols. Ill and IV, shows the same notions projected into the future. Not only do the members of the sect have access to the assemblies of heaven, but the angels come down to those of earth:

War Scroll X.gff; XII. 7ff.

CLEMENT OF ALEXANDRIA

on the whole armour of God . . . for our battle is . . . against the cosmic rulers of this darkness, the spiritual powers of evil in the heavens." άθανα τίζονσιν, accordingly, means "make immortal" by some rite which provided the soul with the necessary purification and information (angelic names and spells) — the armour required for the περιμάχητον πρόσοδον in which, if unarmed, it would be destroyed. For a magical άπαθ αναπαμός introducing the initiate into the company of the gods, see the " Mithras

Liturgy," PGM 1X^.647^741,747,771; also XXIIb.24, 'προσευχή 'Ιακώβ; etc.

That such a magical technique for ascent into the heavens had developed in some Jewish circles during the first century b.c. is also shown by the fact that this date is the latest possible one for the common ancestor of the Hekalot literature and the "Mithras Liturgy" cited above. Scholem in Gnosticism 4.1Ϊ argued that the common ancestor of these traditions must be dated not later than the early second century a.d. and Nock (review of Scholem's Gnosticism) accepted this argument. In my

Observations 1 55fT, I brought evidence for the practice of the technique in the mid-first century a.d. (Col. 2.18) and for its relation to the Essenes; but I did not see clearly its role in early Christianity.17 Now Strugnell (

Angelic Liturgy) has published a text full of the terminology and constructions of the Hekalot hymns. The second of his fragments in particular is rich in such parallels and, to judge from them, describes the wonders of the heavens which the initiate, in his ascent, was expected to see.

This text proves the tradition represented by the Hekalot and the "Mithras Liturgy" was in development at Qumran during the first century b.c., from which Strugnell would date two of the four MSS he used. Consequently there is nothing unlikely in the supposition that Jesus may have practiced such a technique for ascent to the heavens. Moreover, a number of considerations indicate that he did so.

2. INDIRECT EVIDENCE OF JESUS' PRACTICE

First of all, such an ascent was the goal anticipated by much early Christian teaching. We saw above (pp. 203Q that for the Gospels the kingdom of God, the 17. I had not noticed the evidence for Jesus' ascent, and was led by άρπαγέντα (II Cor. 12.2) to think Paul's ascent an involuntary ecstatic experience, peculiar to him. (I neglected the indication that his opponents claimed similar experiences.) Also, I drew a sharp distinction between ascent and instances when the initiate remained on earth but saw heavenly beings or saw the heavens opened (as did ben Zakkai and his disciples, B. Hagigah 14b and parallels, on which Neusner, Life 97ff; cf. Mk. 1 . 10).

But the connection between vision and ascent was very close (e.g., Apoc. 4. iff), and I doubt that such hallucinations can be sharply distinguished. As for άρπαγώτα, I wonder if it may not be an indication of what Paul experienced in his involuntary conversion. (The account in Acts is exoteric and may be fictitious.) Baptism as administered by Jesus was presumably a vivid, hypnotic experience of ascent.

The stages by which it passed from this to a sacramental fiction are unknown and probably differed in each different chain of tradition, as different disciples had more or less of Jesus' hypnotic power. To Paul's churches the gift of the spirit (in baptism?) seems still to have been a matter of vivid experience; but from the way Paul had to remind the members of their death, resurrection, and ascension with Jesus, it would seem that these had already become things which baptism "was supposed to do," rather than experiences actually produced. Within half a century after Paul even the gift of the spirit had probably dwindled to a legal fiction for most of the "orthodox." The gnostics still strove for actual experience, and some may have succeeded in maintaining it well into the second century — hence came some of the charges that they practiced magic.

THE BACKGROUND

sphere of divine rule, is κατ' εξοχ ήν in the heavens. Accordingly, it is to the heavens that Jesus is represented as ascending, at or after his resurrection (Col. 3.1; Jn.

20.17; Acts 1.9; Lk. 24.51 ABC; etc.; Paul probably equated the resurrection with the ascension — Weiss, Christianity 84 — Jn. and Lk. distinguished them). And it is to the heavens that many Christians hope to ascend, either at the End (I Thess.

4.17; Lk. 12. 8f and parallel; Jn. 14. 2f, cf. 6.38; 16.28) or immediately after death

(Lk. 6.23; 12.33 and parallels; 16.22; 23.43 — cf. II Cor. 12.4 — Mk. 10.21 and parallels; II Cor. 5.1-10; Phil. 3.20; Col. 1.5; II Tim. 4.18; I Pet. 1.4; Apoc. 3.21,

7.9-17; etc.) — or even, somehow, in this life (Heb. 12. 22f). Jesus, ascending thither, went as the leader of his followers and showed them the way (Heb. 2.10; Jn. 14. 1-5).

This orientation of the sect might derive from its founder.

In the second place, it appears that claims to have ascended into the heavens were frequently made by Jesus' followers and played an important role in their competition for prestige and authority. One example of such a claim has come down to us — that of Paul in II Cor. 12. 1-5. But from the context there it appears that Paul's claim was not unique (Schmithals, Gnosis I74ff). It was made as vindication of his apostolic status in answer to the claims of his competitors, who presumably were alleging their ascents as proofs of their superiority to Paul (cf. n.2iff; 12. 1 iff; I Cor. 9. iff).

These opponents were "Hebrews, Israelites, seed of Abraham," and ύπερλίαν απόστολοι; therefore it is not implausible that their claims went back to Palestinian

Christian circles. Such claims, put in the mouth of Simon Magus, are repeatedly attacked by the Clementina ( Horn . XVII. 14, 19, etc.). Paul's opponent in Colossians also claimed to have ascended into the heavens, and linked his claims with observance of the OT holy days (Col. 2.18, on which my Observations 156). There may be a touch of polemic against similar teachings in Rom. 10.6. The author of the Apocalypse

(4. iff) claimed to have been taken up into heaven, and to report what he had heard and seen there. II Peter 1.11 recommends that its recipients practice various virtues because ούτως . . . πλουσίως επιχορηγηθησεται ύμιν η είσοδος εις την αιώνιον βασιλείαν.

Here πλουσίως has its normal meaning, "abundantly, plentifully," with reference to repeated entries (contrast Windisch-Preisker, Briefe, ad loc. ; they were forced to make of it "ein glanzender Einzug," although the proof passage they cite from Philo, De vita contemplativa 35, shows the normal reference to plurality). Aphraates,

Demonstratio IV.5, says that Jacob saw in his vision (Gen. 28.1 iff) the gate of heaven, which gate is Christ, and the ladder to heaven, which is "the mystery of {instituted by?) our saviour, by which righteous men ascend from the lower {world) to the {world) above." Another interpretation, preferred by Aphraates (ibid. 6), took the ladder as the cross. The notion that "the mystery of our saviour" means a rite or technique instituted by him is made more likely by the fact that in the contemporary

Hekalot material the technique for ascent to the heavens is also described as a ladder.

In the third place, the supposition that these claims by Jesus' followers reflect some similar claim by Jesus himself is thoroughly in accord with the magical character of Jesus' career, as set forth in the preceding section. The Christian story of Simon

Magus charges Simon with magical ascent to the heavens; the Jewish story of Jesus makes the same charge against Jesus (Actus Petri cum Simone 32; Krauss, Leben 43;

CLEMENT OF ALEXANDRIA

etc.). Eliade, Shamanism , has shown ( passim ) that the claim to have ascended to the heavens and thereby attained supernatural powers is characteristic of a type of magician found all over Asia and the Americas. This type of magician we now begin to find along the Palestine-Syria coast, and soon elsewhere in the Roman empire.

The Vita Apollonii represents ascent to the heavens as the greatest achievement of the Indian magicians and the ultimate triumph of Apollonius (VI. 11 end; VIII. 30 end). In Lucian, Philopseudes 13, it is the conclusive proof by which the (pseudo) sceptic claims that a Hyperborean magician (from the territory of the shamans) overcame his doubts. The magical papyri represent it as the means to achieve magical powers, the reward of their achievement, and the ultimate goal of the magi cian's career {PGM IV.537flf; I.n8ff; XXXIV.yf; I.i78ff).

In the fourth place, the supposition that Jesus practiced some such technique for ascent enables us to explain both the secret character of his baptism and the way he got his disciples into the kingdom of God. Both the Hekalot and the "Mithras

Liturgy" give directions for performing the ascent with a companion or assistant, and provision for a boy as assistant is very frequent in the magical papyri (e.g.,

PGM IV, passim; DMP , passim). Most often the boy is the medium; the magician causes him to see the gods. This technique lent itself for development into an initiation, a "mystery of the kingdom of God," by which chosen disciples could be made to believe they had literally "entered the kingdom." Hence the claim to have experi enced such an ascent appears as accreditation of the apostolate and as authority for teaching, both in Paul and in his Judaizing and Jewish opponents. And how else is this fact to be explained? What authority other than Jesus was both early enough and important enough to give both Paul and his opponents not only the odd notion that they could ascend to the heavens, but the amazing one that such an ascent was necessary to make them apostles of Jesus?

In the fifth place, Paul's interpretation of baptism as a means of ascent to the heavens (Col. 2.gfT; above, pp. 2i4ff) has to be explained. Acts 9.18 reports that Paul found baptism an established practice in the church in Damascus about fourteen years before his correspondence with the Corinthians (II Cor. 12.2) — that is, within half a dozen years of Jesus' death. Presumably his teaching concerning baptism reflected at least the essentials of the teaching given him when he was baptized.

The burden of proof may be left to those who would deny this. They will have to show whence, if not from Jesus, came the Pauline interpretation of baptism as a means of union with Jesus and of ascent to the heavens.

Linally, we have here another contact with shamanism. Eliade {Shamanism 33, 64, 288, etc.) has shown that the traditional scheme of a shamanic initiation involves hallucinations of passion, death, and resurrection prior to the ascent to the heavens.

One of the most striking developments of Paul's, as opposed to the Baptist's, baptism was the notion that it was an initiation ceremony in which the candidate died, was buried, and rose again (above, pp. 2i4ff). Can it be accidental that the followers of a magician credited with shamanic ascent to the heavens came to see in their initiation ceremony the essentials of a shamanic initiation? It can. This particular magician had been caught and put to death, his followers had seen visions of him

THE BACKGROUND

after his death, and these historical accidents may have led to a reinterpretation of the baptism he administered: originally a rite of union with him and participation in his ascent to the heavens, it later came to include also participation in his death, burial, and resurrection. This, at least, has been my assumption thus far; so I have treated all NT references to Jesus' death and resurrection as later reflections of those alleged events. But the theme of deliverance from death, couched in terms suggesting a resurrection, was familiar in the Psalms (e.g., 30.4; 49.16; 86.13) and in the Qpmran Hodayot (II. 32; III. 10, 19; V.6; IX.4ff; X.33f; XVII. 13?) as an expression of deliverance from illness or danger, and, in the Hodayot , from the corrup tion natural to man. It is possible, therefore, that such elements may have figured in Jesus' magical teaching and practice. If so, they would have prepared for the disciples' resurrection experiences. In any event, Paul's interpretation of baptism as death, resurrection, and ascension, and his claim (and the claims of his Christian competitors) to have ascended into the heavens, fit together as pieces of a recognizable pattern. Therefore it seems likely that their common source — the teaching and prac tice of Jesus — contained at least the most important element of the pattern, the ascent to the heavens. (The possibility of a connection with Shamanism has not been disproved by Goldammer's defense of it in Shamanismus.)

3. RECORDS OF JESUS' PRACTICE

The above indirect evidence is supported by some direct evidence from NT passages, chief of which is the story of the transfiguration. This direct evidence is not conclusive, but conclusive evidence about a secret magical practice is hardly to be expected. What we have are several puzzling passages which can all be explained by this one hypothesis more satisfactorily than they have hitherto been by diverse hypotheses. a. The transfiguration The transfiguration is one of the most puzzling stories in the synoptics. Scholars have often noticed its resemblance to four other stories: the baptism (Feuillet,

Perspectives) , the scene in Gethsemane (Kenny, Transfiguration ; Burkill, Revelation

241), the resurrection, and the ascension (see above, pp. 149, 165). To these can now be added the initiation story from the longer text of Mk. This complex set of relations has often led to attempts at source analysis (recently Muller, Verklarung) , none of them convincing. In this group of related stories, the Gethsemane scene differs sharply from the rest because of its central section, the agony, and its historical conclusion, the arrest. The first of these differences confirms the supposition that the agony is a homiletic insert filling the gap produced by the disciplina arcani , which prohibited a report of the initiation Jesus was performing that night (see above, pp. 177 and 237). The historical conclusion, on the other hand, indicates that this type of story had some basis in some type of event recurrent in Jesus' life: it is not pure myth. The nature of the event is indicated by the agreements ol the stories :

a preparatory period or ceremony (John's baptism of Jesus, the last supper, the burial), a small group of participants (in the first baptism, only the Baptist and Jesus?),

CLEMENT OF ALEXANDRIA

a revelation (except in the Gethsemane story, where the police arrive instead). In the two fullest accounts — the transfiguration and the ascension — the revelation involves an ascent, a vision of the glorified Messiah, an entering into a cloud. The first of the stories is the account of Jesus' baptism, and the story in the longer text of Mk. states that the thing revealed was "the mystery of the kingdom of God," that is (as we have seen), a baptism, leading to entrance of the heavens. Paul supposed entering into a cloud to be a form of baptism (I Cor. 10.2); neither Alio, I Cor., nor Lietzmann, Korinther, could gues why.

One could make sense of this data by supposing that Jesus, when baptized by John, had some sort of ecstatic experience in which he saw the heavens opened and the spirit took possession of him (Mk. 1.10), and that using the magical discipline of his day he developed his spiritual gift into a technique by which he was able to ascend to the heavens and also to give others the same experience and similar spiritual powers. Evidence for this might be found in the Gospel according to the Hebrews'. "My mother, the holy spirit, seized me by a hair of my head and carried me off to the great mountain, Tabor." Lk. io.iyff: "The demons are subject to us ... I saw Satan falling from the heaven like lightning. Behold. I have given you authority . . . over every power of the enemy . . . But do not rejoice in this, that the spirits are subject to you; rather rejoice that your names are written in the heavens." Jn. 1.5 1 : "You shall see the heaven opened and the angels of God ascending and descending on the Son of Man." It was seen by Burney, Origin 1 1 5fT that the Son of Man is here the ladder of Gen. 28.12, by which ascent to the heavens is made possible. So in Bereshit

Rabbah 68.12 the angels ascend and descend on Jacob, the mediator (cf. Aphraates,

Demonstratio IV.5, quoted in the preceding section). As the first step in this technique

Jesus continued to use the water baptism of the Baptist, since it removed sin and pre sumably also impurity (which was automatically removed by immersion in sufficient running water) and thus made the initiate fit to enter the heavens. What Jesus added we can guess from Paul, the Hekalot , and the "Mithras Liturgy." What the initiates experienced can be guessed also from the partial, remote, and deliberately secretive reflections in our group of stories. These experiences which Jesus' disciples had during his lifetime probably produced and shaped their visions of him after his death. Thus the resurrection and ascension stories are reflections of the transfiguration experiences which were produced in "the mystery of the kingdom of God." (The ancient world saw the development of many such magical rites and mysteries which purported to make the initiate ascend into the heavens; see the material collected by Lewy, Oracles I77ff. Schrade, Ikonographie 87, saw that recollection of initiatory experience might lie behind the story of the transfiguration.) b. Phil. 2.5-1 1 ; I Tim. 3.16; Jn. 3

These three NT passages imply that Jesus had ascended into the heavens before his death. All three of them may be plausibly connected with baptism. Let us con

sider first the hymn quoted by Paul in Phil. 2.6-1 1 (and therefore earlier than Paul's letter) : <( Messiah Jesus)» ev μορφή deov υπάρχων ούχ άρπαγμόν ήγήσατο το είναι Ίσα θβω, άλλα εαυτόν ε'/ceV ωσεν μορφήν Βουλου λαβών , ev ομοιώμ ατι ανθρώπων γβνόμβνος, καί σχήματί

THE BACKGROUND

ευρεθείς ώς άνθρωπος, εταπείνωσεν εαυτόν γενόμενος υπήκοος μόχρι θανάτου, θανάτου δε σταυροΰ· διό καί ό θεός αυτόν ύπερυφωσεν καί εχαρίσατο αύτώ τό όνομα τό υπέρ παν όνομα, tv α εν τω όνόματι Ίησοΰ παν γόνυ κάμφρ επουρανίων καί επίγειων καί καταχθονίων, καί πάσα γλώσσα εξομολογήσηται ότι κύριος Ίησοΰς Χρίστος είς δόξαν θεού πατρός.

The credal character of this is obvious and its connection with baptism not unlikely.

It has commonly been understood (e.g., by Kasemann, Analyse', Jervell, Imago 2i2f,

227ff) by reference to Col. i, Jn. i, Heb. i, etc., as beginning with an account of the incarnation. But this neglects the fact that it is an account of what was done by Jesus, not by "the Son" (Col. and Heb.) or "the Logos" (Jn.). Nor can it be said that "Jesus" here is a mere slip of the pen by Paul, who carelessly attributed to Jesus a hymn celebrating some pre-existent entity. On the contrary, as is obvious from the conclusion (and as Kasemann has shown at great length) the hymn is a celebration of Jesus. Moreover, Paul expresses similar ideas about Jesus in II Cor.

8.9 : δι' υμάς επτώχευσεν πλούσιος ών. Of course Paul believed in a pre-existent Messiah, but that is not in question here. In both these passages Paul is proposing an example which his disciples can and should follow— the example of their fellow man, Jesus.

Therefore, we must ask: when was Jesus in the form (μορφή) of God and either "as God" (Ίσα θεω) or able "to grasp at equality with God," and when could he have voluntarily relinquished this glory and taken the form of a minister (δούλος =

"DS?) of God — specifically, of a man?18 The answer is: after his ascent to the heavens. After the Helios initiation, quoted above in section C, the initiate is described as ίσοθεου φυσεως κυριευσ ας, and he gives thanks with the words συνεστάθην σου τή ιερά μορφή, εδυναμώθην τω ίερώ σου όνόματι (PGM IV.216— 221 ; for parallels from rabbinic literature and the relation to Phil. 2.6ff see my discussion in Image 478ff).

Having seen Christian baptism as a magical union of the initiate with Jesus and therefore a magical participation in Jesus' ascent, we can now see the force of Phil.

2.5 as an argument for humility and good behavior (which it is in its context, cf.

2.2ff; 2.i2ff). Paul here, as so often, is arguing against a libertine interpretation of the Gospel. When he says τούτο φρονείτε εν ύμΐν ό καί εν Χριστώ Ίησου he concedes that the Philippians have, as a result of baptism, been, like Jesus, εν μορφή θεοΰ.

But they should not attempt, as would the libertines, τό είναι Ίσα θεώ. On the contrary they should now — like Jesus — come down to earth and make themselves serviceable and obedient to their fellow Christians (among others, to Paul).

Moreover, in this instance we can specify the apocalyptic tradition by which the magical practice was interpreted : it is that of Enoch. Enoch began his apocalyptic career as a holy man who was taken up to heaven to see the rewards and punishments reserved for the righteous and wicked, then sent back to earth for a one-year ministry

18. The specification was necessary because the angels are also δοΰλοι του θεοΰ, Apoc. 19.10. On the other hand, no significance whatever can be attached to the distinctions of meaning to be found between μορφή, ομοίωμα, and σχήμα. This is devotional poetry, not philosophic prose. (And even for philosophic prose, see the observations of Wolfson on the laxity of Philo's usage, Philo 1. 102.) In the hymn paraphrased by Paul in Philippians the choice of words was probably determined more by metrical considerations than by those of content, σχήμα and ομοίωμα probably represent Aramaic

terms which were once equated, since the verse in which they stand still shows its original parallelismus membrorum.

CLEMENT OF ALEXANDRIA

to warn men of the judgment (= rule, kingdom) to come: I Enoch 81.5). In 71.11 it is mentioned that he was transfigured when taken on high. This theme is developed in II Enoch, where his transfiguration makes him look exactly like an angel (22.10 — this could be ίσα θεω or ev μορφή θβοΰ, since angels are often called "gods" in Jewish tradition; e.g., Image 477; Kasemann's statement that ίσα θβω could not come from Judaism is mistaken). This time, when Enoch was sent down on his saving mission to men (eh. 36), he had to be changed back to a form men could bear to see (eh. 37).

II Enoch probably dates from sometime in the late first century (Scholem, Gnosticism

17) and is therefore approximately contemporary with the Ascension of Isaiah where, as in Phil. 2, we have the Enoch story adapted to Christianity. In the Ascension of Isaiah , the prophet is transfigured on the way up and becomes like an angel (7.25;

9.30). Then the descent of the Lord is foretold: it involves a whole series of trans formations, first to the likeness of the angels of the lower heaven and eventually into human form (eh. 10). Finally Isaiah is ordered to return into his garment of flesh (= σχήμα άνθρωπον ) and go back to earth (11.35). He is to be a martyr for God and for the salvation of men ( υπήκοος μέχρι θανάτου). Last of all comes III Enoch (closely connected with the Hekalot literature and its doctrine of magical ascent). Here Enoch is transformed into the angel Metatron and is given the name

"Yahweh" (eh. 12, cf. 15), as is Jesus in Phil. 2.9-1 1 ("Yahweh" = κύριος); and all the powers of heaven fall prostrate before him (eh. 14), as they do before Jesus in Philippians. These parallels suffice to show from what tradition the hymn in Philippians has come. The parallels from the Hermetica to which Jervell, Imago

228IT, appeals are less similar in content and more remote from Palestinian tradition.

Here again the closest and earliest parallels to Paul's ideas about Jesus are to be found in Palestinian material. Such parallels establish a strong presumption that Paul got these ideas from Jesus' Palestinian followers, who got them from Jesus himself.

These parallels also make it probable that the similar hymn in I Tim. 3.16, ος έφανερώθη iv σαρκί, έδικαιώθη ev πνβυματι, ωφθη άγγέλοις, έκηρυχθη ev έθνεσιν , e ev κόσμω, aveX ήμφθη ev δόξτ), reflects the same tradition: the gift of the spirit is immediately followed by the first ascension, then comes the ministry on earth, and finally the second ascension in glory. This hymn shows the "adoptionist Christology"

(deriving from the historical tradition that Jesus' messianic delusion began with his baptism) combined with the "pre-existent Christology" which eventually won out.

The two theories can be reconciled in any number of ways; their combination may go back to Jesus himself: he would not have been the only man in antiquity to think himself the son or incarnation of a supernatural power (Wetter, Sohn eh. 1).

This interpretation of Phil. 2.6-11 and I Tim. 3.16 is strengthened by the fact that the same outline of Jesus' career appears more explicitly in Jn. 3 where the connection with baptism is indubitable. After having declared that to enter the kingdom of God one must be reborn of water and of the spirit (verses 3ff — for the rebirth imagery see CH XIII and Festugiere's notes ad loc.), Jesus goes on to describe the magical powers of one who possesses the spirit: Like the spirit itself he can go about invisible; his motions cannot be traced. (So PGM I.222fT; 247flf; etc.; the

THE BACKGROUND

superstition goes back to Plato's ring of Gyges, Republic 359k John probably meant this both symbolically and literally; his Jesus could walk on water, 6.19, and through crowds of men who wished to seize him, 7.44, 8.59, 10.39; and through locked doors,

20.19.) Nicodemus' request for an explanation suggests disbelief (verse 9). Jesus replies (ioff) that all this is well known in Jewish tradition (where magical practices were long established) and that he is speaking from personal experience (ο εωράκαμεν μαρτυροΰμεν) . He then goes on, el τα επίγεια είπον ύμΐν καί ού πιστεύετε, πώς εάν είπω υμΐν τα επουράνια πιστευσετε ; καί ονόείς άνα βεβηκεν εις τον ουρανόν εΐ μη 6 εκ τον ονρανοΰ καταβάς , ό υιός του ανθρώπου. Then comes a reference to the Son's future crucifixion and an account of his role as a means of salvation. This lecture on baptism is fol lowed by the report that Jesus went forth and baptized (verse 22), which leads to the Baptist's "testimony" to Jesus (27ff): Jesus is the bridegroom who has the bride (sc. the spirit, Wetter, Sohn 54); he comes from "above," that is, from God (= "the heaven"), and is therefore above all and ο εώρακεν καί ηκουσεν , τούτο μαρτυρεί (verse

32)· Here Barrett (on verse 13) takes καί ούδείς άνα βεβηκεν, κ.τ.λ., as a comment made by the Church, while Bultmann ( Johannes 108) sees it in a prophecy of the resurrection and ascension — one of the "heavenly things" Nicodemus could not believe. But the plain sense of the passage is: "You do not believe what I tell you of earthly things

<(like the miraculous powers of the baptized), how will you believe if I tell you of heavenly things {the ascent into the kingdom)? Yet {I am the only one qualified to declare such things, for) no one has ascended into the heaven except he who {first) came down from heaven, the Son of Man {that is, I)." So, again, verse 32. Here we have, as in I Tim. 3.16, the combination of "adoptionist" and "pre-existent"

Christologies. For John, Jesus is the incarnation of the pre-existent logos. But this does not prevent John from preserving and reworking material which has come to him from an earlier and more historical tradition, and to such material we owe this recollection that Jesus in his lifetime claimed to have gone up to heaven and to speak of it from firsthand knowledge. (Contrast Sidebottom's contorted attempt to explain away the reference of the ascent.) Parallels to Jn. 3.3 in the Clementine

Homilies XI. 26. 2 and Macarius (.

JVeue Homilien XVI. 3, p. 83) show such agreements against Jn. that Quispel ( Syrian Thomas 230Q argues that this material was handed down independently by a Jewish-Christian Gospel tradition akin to that of the Gospel of Thomas. One of the most important characteristics of this tradition, in Quispel's opinion, was the belief "that eschatology has been realized here and now and that this has implications for marriage and possessions" (p. 235). The possible relevance of this will become apparent later, when we speak of the libertine side of early Christianity.

Other references to the tradition of Jesus' ascent are probably to be found in Jn.

6.38,42,58,62, since it is this tradition — that Jesus had been taken up into heaven, transformed into a supernatural being, and sent back into the world as the messenger of the Father — which explains why the Jews both do and do not know whence he

comes: they know his earthly origin but not his heavenly mission. The same tradition, again, lies behind 10.36, where Jesus speaks of himself as one "whom the Father

CLEMENT OF ALEXANDRIA

hallowed and sent into the world." This description fits a figure like Enoch, who was first a man, then was taken up into heaven, hallowed, and sent back to earth. It does not fit the Logos, which from the beginning "was God" (i.i) and therefore could never have been hallowed. The commentators, from St. Cyril of Alexandria down to Bultmann and Barrett, who try to avoid the difficulty by equating sanctifica tion with commission, probably understand the verse as the author intended it to be understood. But the Johannine context imposes on the words a meaning which they ordinarily would not carry.

Finally, in Hebrews the secret doctrine to which the believers should go on (leaving behind the elementary, exoteric teaching — 6. iff) is that of Jesus' ascent to the heavens (9.1 iff), which enables the baptized to follow him thither (io.igff). Here, as in Paul, the doctrine has been reinterpreted in the light of the passion — Jesus has been made the high priest bearing the sacrifice of his own blood into the heavenly adyton; but this reinterpretation may disguise a teaching rooted in Jesus' magical practice.

E. The rite liberated its recipients from the Mosaic law We turn now to the fifth of the traits of Pauline baptism which were listed as probably derived from Jesus. This is the fact that it results in liberty from the Mosaic

Law. Paul expresses this as a consequence of the death of Jesus: By his death Jesus satisfied the demands of the Law, so that it had no further claim on him; the baptized, being united with him, are also beyond its claims (e.g., Rom. 7.4). There has been much speculation as to how Paul — reportedly a Pharisee — came to hold such a theory, so alien to his training and his moralizing temperament and therefore probably not his own invention. Perhaps the most brilliant presentation of the prob lem remains that by Machen, Origin , who argues that Paul got the main ideas by tradition from Jesus. Here we shall begin by distinguishing between Paul's explana tion of why baptism results in liberty from the Law, and the mere teaching that it does so. Paul's explanation presupposes Jesus' death and therefore can hardly go back to Jesus himself. But the mere teaching makes no presupposition. Therefore it might be supposed to have come from Jesus if there were good evidence of his belief that he and his disciples had been freed from the requirements of the Law.

We should then suppose that his freedom began with his identification with "the Son" as a result of his possession by the spirit at his baptism, and that the freedom of his disciples resulted from their identification with him through their possession by his spirit, in the baptism which he administered (cf. II Kgs. 2.gff, where Elisha gets a double portion of Elijah's spirit as a result of seeing him taken up to heaven in a chariot of fire and/or a whirlwind). It would be plausible also to connect both

Jesus' and his disciples' possession by the spirit and consequent freedom from the Law with their entrance of the kingdom, conceived as a passage from this world or age (subject to the Law) to the coming world or age (of liberty). A century later

Rabbi Elisha ben Abuyah was believed to have learned how to enter the paradise

THE BACKGROUND

in the heavens, and this was thought the cause of his throwing over the Law and becoming a libertine (B. Hagigah I4bff and parallels. Like Jesus, even after he had become a libertine he continued to be consulted as an authority on legal questions, ibid., I5a-b).

The evidence on Jesus' attitude toward law is complex. Here all that can be done is to indicate the major elements. (See further Jesus' Attitude.)

First, the law of the state: Jesus was condemned and executed for criminal sedition; he was arrested at a nocturnal meeting (to which at least one of his followers had come armed, Mk. 14.47 and parallels, cf. Lk. 22.38). Sometime before his arrest he had created a disturbance in the Temple markets (Mk. 1 1 . 1 5f and parallels), and he was accused of plotting to destroy the Temple (Mk. 14.58 and parallel; Jn. 2.19).

As to his teaching about taxes, we have contrary reports: Mk. 12.17 and parallels

(render unto Caesar); Mt. I7.26f (his followers are free but may pay if they wish, from policy). The charge that he forbade payment of taxes appears in one report of his trial, Lk. 23.2.

Second, the Mosaic Law: The evidence is again contradictory. First there are a series of sayings implying that the Mosaic Law is still in force. These are notoriously conspicuous in Mt. (5.17,20; 23.2; etc.) but they occur also in Q and Mk. — Mt.

5. 18 II Lk. 16. 1 7 — it is easier for heaven and earth to pass away than for one iota to fall from the Law. Mt. 23.23 || Lk. 11.42: one should do justice and love mercy and also tithe. Mk. 10.19: obedience to the ten commandments is the way of life; cf. Mk.

12.34.

The same implication — that the Mosaic Law is still in force — is to be seen in a series of stories and sayings in which Jesus interprets one or another commandment either more or less strictly than do his contemporaries; for example, the stories of preparing food and healing on the Sabbath, Mk. 2.23ff; 3. iff; Lk. i3.ioff; 14. iff;

Jn. 5. iff; 9. iff. In most of these Jesus defends his action by arguments from the OT or from precedents in Jewish tradition. These arguments presuppose the validity of the Mosaic Law which they interpret; thus Mk. 2.25ff; Mt. 12.5,11; Lk. i3.ioff;

Jn7.22. The same validity is presupposed by most of the material in Mt. 5 and 23 and parallels. This large and clear body of evidence determines the interpretation of what otherwise might be dubious cases, where Jesus' exegesis is so drastic as to practically annul a provision of the Mosaic Law; e.g., Mk. 10. iff; Mt. 5.38; "Jn."

8. iff. These are to be understood as corrections of detail ( tikkunim ) which do not call into question the validity of the general system. (All this is familiar and is documented in Strack-Billerbeck and Bonsirven, Textes.)

On the other hand, there is an important series of sayings in which the coming of the kingdom in Jesus and his disciples is represented as the beginning of a new age, sharply distinguished from the old age of the Law. "The Law and the prophets were until John, from then on the kingdom of God is proclaimed" and is available to those who will use violence (Mt. 1 1 . 1 2f || Lk. 16.16). Consequently, the least

in the kingdom is greater than Jn. (Mt. 11.11 || Lk. 7.28). The new kingdom is the new garment which is not to be cut to patch up the antiquated fabric of Judaism

(Lk. 5.36) ; it is the new wine not to be put in the old bottles (Mk. 2.22). Therefore

CLEMENT OF ALEXANDRIA

the Son of Man is lord of the Sabbath (Mk. 2.28) and has authority on earth to forgive sins (Mk. 2. iff and parallels; Lk. 7 .47ff ) ; and his companions, as he celebrates his marriage with the spirit, may not fast (Mk. 2.19). As opposed to the Baptist, who represented the Law and therefore came "in the way of righteousness" — that is, asceticism (Mt. 11.18; 21.32), he comes eating and drinking and is called a gluttonous man and a winebibber, a friend of publicans and sinners (Mt. 1 1.19 || Lk.

7.34), in whose homes he is a frequent guest (Mk. 2.15 and parallels; Lk. 15. if;

1 9. iff). His yoke, by contrast to that of the law, is light (Mt. ii.28ff; Acts 15.10).

As John said, the Law came by Moses, grace by Jesus (1.17). With the coming of grace came a new commandment, to love one another (Jn. 13.34; 15.12; Mt. 5.4417).

Accordingly, worship at Gerizim and Jerusalem was to be replaced by worship in spirit and in truth (Jn. 4. 2 iff). For the Son had now revealed to his chosen the hitherto unknown Father (Mt. 11.27 II Lk. 10.22 || Jn. 1. 18), and it was at last possible to assure the initiate, "You shall know the truth, and the truth will make you free"

(Jn. 8.32). "Free" from what? The saying is directed "to the Jews" (8.31), so the implication is, "free from the Law."

Between these opposing bodies of material there is a large no-man's-land of sayings which can be interpreted one way or the other— either as expressions of a liberal legal position, or as manifestations of the freedom of the new age. The neglect of washing before meals is a recurrent example (Mk. 7.5 and parallel; Lk. 11.38). But in spite of this (and of other complicating factors) the difference of the two main bodies of tradition is clear. It is also clear that both are so widely represented in the preserved documents that neither can safely be dismissed as secondary. No doubt it was true (as I argued in Elements) that the rapprochement between the Jerusalem church and Pharisaism in the forties and thereafter contributed not a little to the reformation of Jesus. And the legal arguments to justify the sabbath healings are separable and perhaps secondary. But the bulk of the legalistic material, and the consistency with which it appears in all the sources, rule out any attempt to eliminate it from the primary evidence. On the other hand, the sayings on the presence of the kingdom are now generally recognized as reflections of the most peculiar and original element in Jesus' teaching.

What is needed, therefore, is some explanation of the coexistence of these two bodies of material. And from what we have seen above, the explanation can be supplied. The legalistic material represents Jesus' exoteric teaching. For "those without" (Mk. 4.1 1) the law was still binding, and for them Jesus interpreted it as did the lawyers of his time, holding with more lenient opinions in some instances, with stricter ones in others. (Such variation appears in the opinions of every ancient rabbi.)19 His secret teaching was only for those to whom the mystery of the kingdom

19. How it came about that Jesus was consulted as a legal authority (e.g., Mk. 10.2; Lk. 12.13) we do not know. It is not impossible that he had some legal training. As the Hekalot tracts, the stories of Yohanan ben Zakkai (on which Neusner, Life 97, etc.), and the magical material newly discovered by Margalioth demonstrate, the halakic tradition, even in rabbinic Judaism, has close connections with the study of magic, including the practice of ascents to the heavens. That these could be combined with libertinism, too, and that the successful magician and notorious libertine could remain a great legal authority, is shown in the case of Rabbi Elisha ben Abuyah. On the other hand, anyone in first-century

25O

THE BACKGROUND

had been given (ibid.). He had no intention of giving that which was holy to the dogs, or of casting his pearls before swine (Mt. 7.6). The contradictions of the present

Gospels may result from seepage of secret material into originally exoteric texts.

Of this we should now have a further example in the additions of the longer text to Mk. More evidence might be seen in the fact that material suggesting the presence of the kingdom, as a radically new regime exempt from the laws of the former age, is more prominent in Q, and L than in Mk. — it was added to Mk. by Matthew and Luke. Later additions may be found in Jn. 8. iff and in the western text (D) at the end of Lk. 6.4: "On the same day, seeing a man working on the sabbath, he said to him, ' Man, if you know what you are doing, you are blessed, but if you do not know, you are accursed and a transgressor of the Law.'"

More of the esoteric teaching is found in the epistles of Paul, the oldest Christian documents and those most surely written for reading within the closed circles of the churches. Within those circles there were still distinctions in degree of initiation

(I Cor. 3. iff) — esoterism is rarely content with a simple inside-outside contrast, but loves to elaborate secrets within secrets to the thirty-third degree. Nevertheless,

Paul enables us to glimpse the true beliefs of the congregations to which he writes, and he is to be preferred, as a source for early Christian thought, to the later, com paratively exoteric Gospels.

XI. Consequences of Jesus' baptismal practice Through the preceding studies of the relations of Jesus' work to that of the Baptist and of Paul, we have arrived at a definition of "the mystery of the kingdom of God" :

It was a baptism administered by Jesus to chosen disciples, singly, and by night.

In this baptism the disciple was united with Jesus. The union may have been physical

(see above, commentary on III. 13 and pp. 185!— there is no telling how far sym bolism went in Jesus' rite), but the essential thing was that the disciple was possessed by Jesus' spirit. One with Jesus, he participated in Jesus' ascent into the heavens; he entered the kingdom of God and was thereby set free from the laws ordained for and in the lower world.

This understanding of the mystery of the kingdom enables us to understand in the history of the early Church a number of problems which hitherto have gone either without solution or without recognition. The failure to recognize and the inability to solve them have in part been due to the influence of Acts, a partizan document often shown to be incomplete (it neither reports the foundations of the churches of Rome and Alexandria nor the survival of Christianity in Galilee,

Palestine who had messianic pretensions would have had to give decisions on legal questions. In Jesus' Attitude I discuss the whole problem further and propose (p. 244) a threefold division of Jesus' hearers: ordinary Jews, for whom the Mosaic Law is still binding, those who would be perfect, who are still subject to the Law but with additional requirements and also with special exemptions, and those who have been admitted to the kingdom, who are free from the Law.

CLEMENT OF ALEXANDRIA

cf. Lohmeyer, Galilaa; etc.), confused (5.19 is a doublet of i2.7ff; Peter's miracles are doublets of Jesus' ; the chronology of Paul's career is hardly to be reconciled with Gal. 1 and 2), and misrepresentative (it minimizes doctrinal differences in order to present a single, original, united Christian Church, of which Paul was the fully approved representative — cf. Weiss, Christianity 260; Brandon, Fall 2o8ff; on the Hellenists, Cullmann, Significance ; Simon, Stephen , vs. Haenchen, Apostelgeschichte, on 6.1). But the blame cannot be laid entirely on Acts. Bauer, Rechtglaubigkeit , has shown that the NT as a whole is a partizan collection, made to present and support the views of that party which became predominant within the Church in the late second century and finally triumphed in the third. Accordingly, the picture of early

Christianity given by the JVT has constantly to be supplemented by pictures of the parties it opposes , and has frequently to be explained by reference to factors it attempts to conceal. One of these concealed factors is Jesus' practice of baptism as "the mystery of the king dom." Therefore a number of consequences of this practice appear in the history of early Christianity as unexplained phenomena or are "explained" by stories of miracles. These can now be explained historically by recognition of their source in Jesus' practice.

A. The coming of the spirit Most critical scholarship has uncritically taken for granted that "the spirit" — that is, the schizophrenic behavior — just "came," and has not tried to explain its coming.

Acts "explained" it as a miracle (Pentecost). But the coming of the spirit presupposes a group with peculiar preparation. The holy spirit did not descend on the disciples of Rabbi Akiba after he was executed for treason ; at most they heard a heavenly voice (B. Berakot 61b). But Jesus' disciples had been prepared for the phenomena of group possession by their experience of individual possession in his baptism. It must be remembered that he had a peculiar attraction for and power over schizo phrenics. Hence his exorcisms and his following of "women who had been cured of evil spirits," Lk. 8.2. The stories of his disciples' sudden, total abandonment of their ordinary lives, to follow him— Mk. i.i6ff; 2.14; Jn. 1.43; cf. Lk. 9-59ff — probably reflect the same power and explain not only the disciples' suitability for possession in baptism, but also their visions of the resurrected Lord. These same psychological characteristics were perpetuated in the Pauline churches (above, p. 215).

This explains also why the spirit is identified with Jesus, as it is in Paul (I Cor.

6.18; 15.45; II Cor. 3. 1 7f) . Jesus had originally been the source of it. When the type of personality disturbance originally connected with him recurred as a group phenomenon in the circle of those who had depended on him for it, he was of course supposed to be the cause. This, however, was a historical inference, and as time went on and personal memories of Jesus faded, the spirit became an independent personality and pushed Jesus aside. This can be seen happening already in Paul. For, unlike

Jesus, the spirit was alive and present: it spoke to the churches (Apoc. 2.7,11,17, etc.), did miracles, and was the source of knowledge and wisdom (I Cor. I2.8ff)

THE BACKGROUND

and the guide of private life (Gal. 5.18,22, etc.). Therefore, as the churches grew, the importance of the present, active spirit grew with them, while its connection with Jesus receded into the background — a memory for the few he had initiated, a dogma for the many new converts.

Accordingly, for Acts the spirit is something Jesus gives after his death, resurrection, and ascension. The prophecy of Jesus' baptizing by the spirit, which Mk. 1.8 put into the mouth of the Baptist, is made by Acts to refer to the beginning of the group experience after Jesus' death: 1.5; 11.16. (Another factor in the account given by Acts is secrecy. Luke's silence about the baptismal teaching of Jesus, like his silence about Paul's theology, can hardly be explained save as deliberate reticence. For further evidence on this point, see Reason .)

Development has gone even further in the latest stratum of Jn., which insists that the spirit was not given until after Jesus' death and ascension (7.39; cf. 14.17,

26; 15.26; 16.7,13; 20.22). The insistence looks like polemic and is perhaps to be explained along the line suggested by Kragerud [Lieb lings j Unger) and Kasemann

( Ketzer ) : John is appealing to the newly given spirit for authorization, against opponents who had better claims than he to human traditions about Jesus. New problems required a new authority, so the development of the doctrine of the Third

Person of the Trinity was under way, though its progress was soon to be slowed.

As the original enthusiasm of the Christian communities wore off, the temporary attacks of schizophrenia which had been the most conspicuous gift of the spirit became rarer. Also, their practical importance presumably declined as the frequent foolishness of the promptings they produced became evident, as the churches developed fixed routines for worship and daily life, and as control of the churches fell more and more into the hands of professional administrators who are not, as a class, much given to ecstasy and have often viewed ecstatics with hostility as possible competitors for authority. From the mid-second century on, therefore, the active guidance of the spirit is characteristic of groups soon thereafter to be declared "heretical."

B. The adoption and development of baptism by the early churches A second problem explained by Jesus' practice of baptism is that posed by the fact that Jesus' followers practiced a baptism of the Baptist's type, yet claimed that it would not only remit sins, but also give the spirit. If Jesus never baptized (and, even more, if he deliberately stopped baptizing) it would be very difficult to explain why his followers took up the practice (above, pages 2ogf). And even if we neglect this difficulty and suppose they did take over the practice from the Baptist's followers, where did they get the notion that baptism gave the spirit, which the Baptist's baptism did not (Acts 19.1-6)? The most probable explanation is that the first element of Jesus' baptismal rite had been an immersion like the Baptist's for remission of sins and also for the purity necessary to approach the heavenly kingdom. (Until

baptism the candidate was still of this world and therefore bound to obey the Law.)

What Jesus had done beyond the Baptist's immersion was something dependent

CLEMENT OF ALEXANDRIA

on his peculiar powers of suggestion. But even if some of the early disciples succeeded in repeating Jesus' rite of possession and ascension, their initiations of a few individuals, one by one, were soon overshadowed by the phenomena of group possession in the meetings of the churches. The spirit now united the believer to the life of the Church, not to the person of Jesus, and the entrance to the kingdom in the heavens was replaced by the entrance of the Church, the kingdom present in this world (above, pp. 202f). Consequently, the developments which Jesus had added to the Baptist's baptism fell into disuse or were preserved as "great mysteries" for more advanced candidates. But the Baptist's baptism was preserved as the necessary ceremony for entrance to the Church. Since the Church was the kingdom of God on earth, entrance required that the initiate be pure and sinless. Since the Church was also the body inhabited by Jesus' spirit (I Cor. I2.i2f), ceremonial entrance of the Church came to be credited with effecting possession of the spirit. This interpretation we find already in Paul, but Acts reflects the tradition that although Christian baptism is normally followed by the gift of the spirit (2.38; 19.5^, nevertheless it need not be so. The coming of the spirit may be delayed for some time after baptism (8. 12ff) or may even precede it (10.47; cf· n.i5ff; Flemington, Baptism 39ff).

C. The libertine tradition in early Christianity A third consequence of Jesus' baptismal practice was libertinism. The Christian

Libertine tradition, because of its early date and wide extent, is best explicable as the consequence of Jesus' teaching that those who have entered the kingdom are free of the law. Something has already been said of this as the source of Pauline theology, and the importance of Paul's doctrine needs no exposition. What has not generally been realized is that Paul as an exponent of Christian liberty was neither unique nor extreme. On the contrary, he was a comparatively conservative and mediating figure, consistently attacking and attacked by representatives of the libertine side. The libertine party or parties, whose tradition derived from Jesus himself, must have been widespread and influential, since evidence of their importance is to be found in almost every book of the NT and in most of the extracanonical Chris tian literature of the first two centuries. However, since all of their own writings have been destroyed, the very existence of the movement has too often been overlooked.

Consequently, it seems worthwhile here to sketch at some length the major elements of the evidence.

I. THE PERSECUTION OF THE EARLY CHURCHES

This problem has been pointed out above (p. 234), but deserves further considera tion.

As reported by Acts, the history of the early church of Jerusalem was a series of

persecutions. But why should Jesus' followers have been persecuted? To believe that the Messiah had come, or to expect that he would soon come again, were not offenses punishable by Jewish law. And a man like Paul would not have taken an active

THE BACKGROUND

part in a persecution (as he himself testifies that he did — Gal. 1.23; I Cor. 15.9) without serious cause.

The concern of the temple authorities to stop the apostles' preaching in Jesus' name because "you wish to bring upon us the blood of this man" (Acts 5.28) is understandable if it means "you are inciting the mob to riot and punish us for handing him over to the Romans." But this can hardly be taken as the Christian message; neither will it explain the persecution by Paul, a Pharisee, who presumably had little concern for the welfare of the Sadducean high priests. Again we are told that the Sadducees were "grieved at their proclaiming in Jesus the resurrection from the dead" (Acts 4.2). If this meant simply "teaching that there would be a resurrection of the dead, and using Jesus as an example," it would not have been grounds for legal action, since the Pharisees also taught the resurrection of the dead and the appeal to an example does not alter the teaching. Perhaps the author wished to represent the Sadducees as acting illegally and thus expressed his hostility to the Sadducees, by contrast to the Pharisees whom he represented as not always unfriendly (so Blass, Acta, ad loc., comparing 5.17,34; 23.6ff). Another possibility, however, is suggested by Paul's warning the Thessalonians against those who taught that the day of the Lord is already here (II Thess. 2.2). If the teaching in Jerusalem was that "the resurrection" had already come "in Jesus," it might have led — as it did in Thessalonica (3-6ff) — to irregularities of behavior which could have occa sioned legal action.

In Acts 6.14 Stephen is charged with having said that Jesus would destroy the temple and abolish (or, change?) the Mosaic Law. Stephen proceeds to deliver a speech which tends to justify these charges, and is lynched. Here again, while his prediction of such events in the future might have moved a Jewish audience to lynch him, the action would have been more understandable had he been teaching that the end was already come and the law already abolished for those in Jesus. It was this latter teaching which almost got Paul lynched in the same place about twenty years later (Acts 21.21,28), and something similar is attributed to Jesus half a dozen years earlier (Mk. 2.10,19,28) in a body of material intended to lead up to and explain the Pharisees' plot against his life (Mk. 2. 1-3.6).

Finally, no explanation at all is offered for the beginning of the persecution under Herod Agrippa I (a.d. 41-44), but it began with the beheading of James, the brother of John (12.2), and even in first-century Palestine capital punishment usually required serious grounds. Acts says the persecution was continued "because it was pleasing to the Jews." But some fifteen years later, when Paul comes to Jerusalem in Acts 21, we find the church on excellent terms with its neighbors and making "tens of thou sands of converts ... all of whom are zealots for the law" (21.20). Perhaps the change was begun by Herod's persecutions, which drove Peter underground and so allowed the leadership of the Jerusalem community to pass to James, Jesus' brother (Acts

12.17— Acts' first mention of James). The question of cause of the persecutions is thus related to the question of the parties in the Jerusalem church, where, it would

seem, the libertine wing was originally important but was expelled by outside pressure and by developments within the Christian group. The leaders of the legalist

CLEMENT OF ALEXANDRIA

party within the church may have asked their friends in other Jewish parties — particularly the Pharisees, who had much influence with Herod Agrippa I — for a little timely persecution to help get rid of their libertine rivals. See further Reason.

2. THE PARTIES IN THE EARLY CHURCHES

The same picture emerges from the preserved evidence about the parties in the Jerusalem Church. This evidence is presumably incomplete, and what is preserved is none too good. Even among the accepted Gospels the lists of the twelve closest followers of Jesus differ slightly; the libertine parties may have appealed to other figures like Salome (above, pp. i8gff) or Glaucias and Theodas (Clement III. 75. 1 5fT) , whose names have been eliminated from the accepted tradition. Trocme, Formation

42 ff, 90-108, etc., sees in Mk. a tradition which emphasized the role of Jesus as a miracle worker, not to say magician, and was hostile to the "respectable" tradition which appealed to Peter and John and represented them as the authoritative apostles.

But we need not speculate as to what has been lost (though we must always remember that much certainly has been) ; there is plenty of evidence in the preserved material for the relation of different figures of the Jerusalem church to different parties.

John is particularly hostile to Jesus' brothers; he goes out of his way to state that they did not believe in Jesus during his lifetime (7.5). But none of the Gospels represents them as part of his following. We may suppose that they came into the movement after his death; as members of the family of an alleged pretender to the throne, they were involved in his disaster whether they liked it or not. The probability therefore is that James (like the other converts made since Jesus' death) had never received Jesus' initiation. His succession to leadership of the Jerusalem church will have marked the triumph of the converts over Jesus' original circle. Significantly, his authority had to be supported by the story that he had seen the risen Lord (I Cor.

15.7) — Paul was not the only one who appealed to posthumous revelations. After James' triumph, the antinomian aspect of Jesus' original teaching was obscured.

Jerusalem became the center of a legalistic interpretation of the Gospel. Repercussions of this in the mission field are seen in Galatians and in Acts 2i.i5ff, from which it is customary to distinguish four positions.

First is the position of the legalists, whom James is trying to pacify and for whose sake he urges Paul to show that he still observes the law (Acts 2i.2off; Gal. 2-3f).

Since this party maintained that the law was still binding, they must have denied either that the kingdom could as yet be entered, or that entrance to it put one beyond the law.

Second, the position of James : the appearance of obeying the law is to be maintained.

This concern for appearance (ρ^Π ΓΡΚΊΰ) is of great importance in Jewish law (e.g., KiVayim III. 5; IX. 2; Shebiit III. 4; Shabbat XIX. 6; Bekorot VII. 3, 5). If Acts 2i.24ff can be trusted, James and Paul were agreed that full observance of the law by gentile converts was not required, and James presumably had few illusions about Paul's practice outside Jerusalem (21.21, cf. Gal. 2.12). What he urged and Paul accepted was an act of "occasional conformity." Gf. Gal. 6.13 and Paul's argument to deter the Galatians, 5.3, which is proof that they were being asked to

THE BACKGROUND

obey only a few conspicuous commandments. Another concern perhaps relevant here is that for living at peace with one's neighbors (DlVttf OVt Shebiit IV.3;

V.g; Gittin V.9; Shekalim 1. 3). But evidence as to rabbinic law in this period is so unreliable that there is no point in trying to define closely the legal positions of the several parties.

Third, the position of Paul: there is no need to appear to obey the law unless apparent disobedience of it would lead you into danger or your fellow Christians into sin (I Cor. 8.gff; 10. 14-31). However, one is free to obey when obedience seems politic (I Cor. g.igff), provided that one does so with clear understanding that obedience and disobedience are of no importance for salvation (I Cor. 8. 7f )

Fourth is a mediating position (?) represented by Peter and Barnabas, who felt no obligation to preserve appearances among the lax but wanted to keep on good terms with James (Gal. 2.11-13).

On this customary analysis, two observations must be made. The first is that James' position practically presupposes that the doctrine about the liberty of those in the kingdom will be kept secret. The legalist party in James' church was probably the source of pronouncements like Mt. 23.2: "The scribes and the Pharisees sit on Moses' seat; all things whatsoever that they say unto you, these do and observe."

Such a party could hardly have lived at peace with another which held that obedience to the law was optional, unless the latter opinion was masked by obedience and kept secret. Similarly, the position of Paul is congenial to secrecy, though it does not require it (cf. Reason). The existence in his churches of members who do not know that the eating of things sacrificed to idols is harmless (I Cor. 8. 7fT) implies that his doctrine of Christian liberty had been kept to an inner group. I Cor. 3.1 indicates that he had teaching reserved for more advanced believers and kept secret from "babes." These "babes" were ignorant of the mysteries revealed by the spirit (I Cor.

2.6-16), of the things to be revealed at the end (13.8-13), and of their own freedom from the law (ibid. ; 8.7). (How such ignorants got into a Pauline church is a question to be noted in passing for its indication of the difference between what Paul preached when he first came to a community and what he taught later.)

This leads to the second observation on the customary catalogue of parties, given above. The catalogue follows Acts and therefore neglects one important factor in the situation, the libertine party or parties to the left of Paul — groups like that against which Paul defends himself in I Cor. 8ff. Acts says nothing of this side of Christianity (save for an occasional veiled reference, like 20.2gff), and its silence is not accidental. One of the author's purposes was to persuade his Roman readers that Christianity was morally admirable and politically innocent (

1 8. 1 qf; 23.29; 25.25; 26.31). Therefore the libertinism, usually scandalous and occasionally criminal, was concealed. Consequently scholars have neglected it, and have treated the references to it in the NT as references to particular abuses in practice or corruptions in doctrine peculiar to the single church in connection with which they are mentioned.

Admittedly, as Koster has remarked ( Hdretiker , sec. 3 end), it is not safe to suppose that every warning against moral turpitude implies the existence of a libertine sect.

Sin occurs also in the most legalistic communities, as Meleager of Gadara happily

CLEMENT OF ALEXANDRIA

observed: εστι καί εν φνχροΐς σάββασι θερμός ' Έρως ( Anthologia Palatina V. 1 60.4)· Nevertheless, the arguments of Jude or I Cor. cannot be explained by individual offenses unconnected with theological theory. And the existence of clear cases of theoretical libertinism requires that other cases be examined for their relevance.

Accordingly, we now turn to these.

3. DIRECT EVIDENCE OF LIBERTINISM

Here is a list of the passages in which polemic against libertinism is fairly clear:

Mt. g.ig. Those who teach others to break the commandments, and are therefore the least in the kingdom, are evidently libertine teachers. This indicates the interpre tation of 5.1 1,13,17 fT, etc., and also of Matthew's additions of δικαιοσύνη to Q sayings like Mt. 5.6 and 6.33 — to say nothing of the emphasis on the need for good works which runs throughout the sermon on the mount.

Mt. 7.13-27. These teachers pretend to observe the law, but their secret teaching is contrary to it. They are false prophets and are conspicuous for their magical powers, displayed in prophecies, exorcisms, and miracles done ' 'in Jesus' name" (cf. Col.

Mk. g.42. Whoever scandalizes one of the "little ones" among the believers would be better off dead. Cf. I Cor. 8.9-12. This may indicate for whom the editor intended the warnings of Mk. 9-43ff.

Lk. 7.36-go. This has been edited. From the premise "he who is forgiven more will love more," the conclusion to be drawn is that the greatest sinners (who have most to be forgiven) are capable of the greatest love. This Luke has replaced by less dangerous doctrine. He has similarly edited the saying "I came not to call the righteous, but sinners" by adding "to repentance" (5.32 || Mk. 2.17b; Mt. 9.13).

However, he preserves a lot of peculiar material which suggests that the kingdom is primarily for sinners (i5.8ff, the lost coin; I5.i2ff, the prodigal son; 16. iff, the unjust steward). This looks like a source. It had some material in common with Q(i5.4ff, the lost sheep || Mt. i8.i2f; 15.10, joy in heaven over one repentant sinner ||

Mt. 18.13 — here again Luke has edited: Mt. has more joy over one repentant sinner than over ninety-nine righteous.) This material is not necessarily libertine (cf. Barnabas

5.9), but is patient of libertine interpretation; and Luke's care to specify that the sinners were to be called " to repentance" is evidence that such interpretation troubled him.

Acts 20.2gf Paul's ex eventu prophecy to the Ephesians of the wolves among the sheep attacks teachers of "twisted things" and is interesting because it is related both to Mt. 7. 1 5ff (supra) and to II Thess., where, as here, Paul is attacking men who live on the congregations and refuse to work (II Thess. 3-6ff). These men should probably be connected with the belief that "the day of the Lord is already here" (II Thess.

2. if). This belief was being foisted on Paul by epistles forged in his name (ibid.).

THE BACKGROUND

Further references to the same theme appear in II Tim. 2.i6ff and perhaps II Peter

3-3f; see below. This may also have been the belief of those in Corinth who denied the resurrection and, it seems, claimed to have received salvation "in this life"

(I Cor. 15.19) ; they, too, were a libertine group (15.32-34). See also below, on Apoc.

Rom. 3.8. "Let us do evil that good may come" looks from the context as if it were said by libertines who claimed that Paul's teaching agreed with theirs; cf. also Rom.

6.1,15-23. Another quotation may be 8.1. In 8.12 the notion that we are not debtors to the flesh appears suddenly and without relevance to what has preceded. Who said we were debtors to the flesh, that Paul should deny it? Mt. 5.25k || Lk. 12.58k

("agree with your creditor quickly") was understood by unnamed teachers, whom Clement opposed, to refer to the body (II.290.10ff); the Carpocratians understood it to refer to the demonic ruler of this world, the creator of the body (Irenaeus:

Harvey, 1. 20. 2 = Stieren, 1. 25. 4; cf. Tertullian, De anima XXXV). Thus, through the first part of Romans the argument often seems to have been framed with reference to libertine arguments; this makes it plausible that the same reference may explain some of the moral counsels in the second half. (They can hardly refer to particular circumstances in the Roman church, which Paul did not know.)

I Cor. 3.2. This is unmistakable; the Corinthians were proud that incest was being practiced in their church. (This anticipates the accusations against Christians reported by the apologists of the next century.) 6.9f, "do not be deceived," indicates the existence of an opposed teaching; and 6.12, "all things are permissible," is probably one of its premises — one Paul is not prepared to deny but attempts to "explain." Another premise, that knowledge (of what?) has freed the initiates from the Law, appears through 8. iff. Cf. Jn. 8.32 : You shall know the truth and the truth shall make you free. Paul finds it necessary, in 9.1, to remind the Corinthians of his own freedom, presumably to equal the claims of his opponents. 10. iff proves that they based their claims on the supposed effect of baptism and communion — no doubt because these rites gave them the spirit, and the spirit had the knowledge

(I Cor. 2.6ff) — but also probably because baptism included resurrection. Hence their denial of a future resurrection (15.12), claim of salvation already in this life

(15.19), and consequent libertinism (15.32-34); cf. above, on Acts 20.29^ and below, on Apoc. 2.

Gal. 5-i3ff. Here it appears that Paul was aware of the libertine interpretation as a danger to which his doctrine laid itself open. This awareness was probably the result of experience, which probably continued and got worse, see above, on Rom.

3.8. That libertinism was an immediate danger to the Galatians is not suggested.

Phil. 3.1 8f. "The enemies of the cross of the Messiah, whose end is destruction, whose god is the belly and whose glory is in their shame" are probably libertines.

Since verse 1 7 seems to take up a new subject, they are not to be identified with

the κατατομή of verses 2-16, that is, those who teach that the observance of the Law is still necessary. "Enemies of the cross" probably refers to docetism (cl. Jn.

CLEMENT OF ALEXANDRIA

1 91 7) 5 " whose god is their belly" may refer to magical practices (cf. Εγγαστρίμυθος) and "whose glory is in their shame" to their pride in shameful actions as signs of liberty (cf. I Cor. 5.2, above). Paul's explanation in the preceding section (3.12), ονχ on ηδη Ελα βον η ηδη τζτελζίωμα ι, may be an anticipation, in arguing against the legalists, of the dangers of the libertine interpretation (cf. Gal. 5.13).

II Thess. 3-6ff. See above, on Acts 20.29b Eph. 3.1-20. This attacks specific opponents. It is shameful even to say what they do in secret (verse 12); they try to deceive the Ephesians and to persuade them that sexual license is not incompatible with the kingdom of God (verses 5f) ; they practice "fruitless works of darkness" (verse 11, nocturnal ceremonies?) in which the Ephesians should not participate, since they have been raised from the dead

(in baptism) to participate in light (11-14). The specific reference is also proved by comparison with the generalizing morality of Col. 3.5-17, which the author of Eph. was reworking.

I Tim. i.igf. Hymenaeus and Alexander, by rejecting conscience, made shipwreck of their faith; consequently (pseudo) Paul gave them over to Satan that they might be taught not to blaspheme. This is presumably directed against their followers, whom it taunts with some of their teachers' misfortunes believed to have been caused by Paul's magic. "Rejecting conscience" is a vague charge, and the rest of the letter seems to be directed against legalists (

1 . 7fT; 4. iff); so this passage would hardly belong in the present list were there not a likelihood that the Hymenaeus here is to be identified with the Hymenaeus in II Tim. 2.16, below. (6.3ff and 6.2of are of uncertain reference; all Christians claimed knowledge and most wanted to get rich.) II Tim. 2.16-4.4. Hymenaeus and Philetus said the resurrection had already taken place (probably magically, in baptism) and therefore did not abstain from un righteousness (verse 18). Their teachings probably provided opportunities for the indulgence of youthful desires (22) and for idle controversies (23). They have been caught in the snare of the devil and are doing his will (26). They are apparently examples of a much larger class whose members have all the vices (3-2fT) and are particularly successful in their libertine mission with women, to whom they promise secret knowledge (3-6f). They are magicians, like Iannes and Iambres (3.8,13), and teach myths (4.4). The teaching that the resurrection has already taken place links these men to the group(s ?) attacked in II Thess. 3.6ff. Acts 20.29, etc· 1 see above.

James 2.14JJ. The argument here is against someone who claims to have faith but does not have good works (14) and evidently thinks good works unnecessary for salvation. Cf. 3.13-18 || Mt. 7.21 (above). The opponent may be Paul, but there is so much evidence for libertine teaching that the identification is dubious.

I Peter 2.11-16. This is mere moral exhortation, not directed against anybody.

But the final warning, that Christian liberty should not be made a cloak for wrong doing, shows that the author was aware of libertinism, though he did not consider it a real and present danger to the group for which he was writing.

THE BACKGROUND

II Peter 1.5-11. The insistence that one must have virtue as well as faith, and conti nence as well as gnosis, is probably aimed at some libertine teacher (attacked in verse 9) and is backed up by the promise that virtue will be rewarded by frequent ecstatic experiences (entries to the kingdom, 11).

II Peter 2-3 and Jude. This tract was evidently popular with the victorious party, since two versions were canonized. The popularity indicates that there was con siderable need for it — hence that the libertine wing of Christianity was strong and widespread. The "false prophets" of II Pet. 2.1 recall Mt. 7. 1 5fT. The denial of Jesus may be docetism (cf. Phil. 3.18; I Jn. 2.22f). The ασέλγεια (2) is the hallmark common to most passages. That the libertines give Christianity a bad reputation antici pates the excuses of the second-century apologists. Their contempt for the cosmic powers (ioff || Jude 8ff) is a new and surprising charge; it suggests that the tract may originally have come from a semilegalist teacher like the one attacked in Col. 2.16.

So does the haggadah (fallen angels, Noah, Sodom, Lot, Michael and the body of Moses, Cain, Balaam, Enoch. The Balaam theme has recently been linked with an early Syrian center of Christianity with which Stephen's speech declaring the abolition of the law, Mt., Simon Magus, and the Nicolai'tans of Acts may also have been connected: Danielou, Etoile 133-136). What the libertines understood physically, to their consequent corruption, may have been sayings about the freedom in the kingdom. Agreement between II Peter and Jude ends with the statement that their opponents are mockers intent on their own desires (II Pet. 3.3; Jude 18). To this

Jude adds that they are ψυχικοί, πνεύμα μη έχοντες; II Peter, that they deny the second coming; the latter may be a different form of the claim that "the day of the Lord is already here"; cf. above, on Acts 20.29.

I Jn. 1.6,8. 1.6 might be ordinary morality; but 1.8, the claim to be sinless, must come from a liar, a fool, or a theologian. The author opted for the first and second of these possibilities, but the third is not thereby excluded. We may suppose that perfectionist, libertine theology was one of the writer's targets. Confirmation for the supposition can be found in 1.10; 2.4, i8f (they are antichrists), 22f (also docetists, see above; Phil. 3.18; II Pet. 2.1; Jude 4); 3.7-10 (they commit sinful actions);

4. iff (they are false prophets, cf. II Pet. 2.1, Mt. 7. 1 5ff) ; 4.5 (they have considerable success in their missionary work). More of the same appears in II Jn. 7.

Apoc. 2.6,1 4f,2off. Nicolai tans are established in Pergamum, hated in Ephesus; they have irregular sexual practices and eat things sacrificed to idols, thus following the teaching of Balaam (see above on II Pet. and Jude). In Thyatira there is a woman whom the author calls Jezebel, who calls herself a prophetess and teaches the same practices. She has been at it for some time and has not taken the opportunity to repent. Her followers claim to know "the deep things of Satan" (or perhaps simply

"the deep things," if "of Satan" be the author's comment). The complaint about eating things offered to idols suggests that these letters, like the tract behind II Pet. and Jude, come from a comparatively legalistic circle.

The above list has been limited to passages of which the purport is reasonably clear. Both discussion of disputed details and bibliographical data have been excluded,

CLEMENT OF ALEXANDRIA

since the most important thing in this evidence is the cumulative force. (For a recent bibliography see Koster, ΓΝΩΜΑΙ 283 n8, and for the variety and importance of different trends in the early churches see the whole article.) The passages listed above show that libertine Christianity was widespread and ancient. It is attested in Ephesus,

Corinth, Thessalonica, Pergamum, and Thyatira, and must have been important in Syria, too (Mt., the tract used by II Pet. and Jude). The connection of the Nicolai'tans of the Apocalypse with the Nicholas of Acts 6.5 is suggested by what is said in Clement (II. 207. i8ff) and asserted by Irenaeus (Harvey, 1. 23 = Stieren, 1. 26. 3); it involves no improbability. Hippolytus, De resurrectione ( Werke , GCS vol. 1 .II, p.

251), says the basis of Nicolas' libertinism was the belief that resurrection occurs in baptism (a belief Hippolytus does not contest). This looks like correct information; cf. above, on Acts 20.2gf; I Cor. 15; II Pet. 2-3. The Nicholas of Acts 6.5 was one of the deacons appointed to appease the Hellenists; and some scholars have supposed that the Hellenists, if a libertine or laxist group, stood closer to Jesus' teachings than did the Hebrews (Conzelmann, Apostelgeschichte 43; cf. the remarks on Jesus by Gartner, Temple i2off). The connection of libertinism with docetism cannot be used to prove the former late; the connection may be secondary. Moreover, the hymn quoted by Paul in Phil. 2.6ff is probably docetic in supposing that Jesus' resumption of humanity after his ascension was in appearance only — pace Kasemann,

Analyse 340 — and the Enoch tradition is certainly so in the Ascension of Isaiah , which is probably first century. Finally, whatever may be thought of such arguments, the fact remains that the libertine tradition is attacked by Mt., Mk., Lk., Acts, Paul

(in five letters), Ephesians, I and II Tim., James, I and II Pet., I and II Jn., Jude, and Apoc.

4. SOURCE AND DISPERSION OF THE LIBERTINE TRADITION

The fact that the libertine tradition was so early so widespread is evidence that it derives from Jesus' baptismal practice. To try to explain all of it as resultant from misunderstanding of Paul is implausible. Paul protests that his name and his sayings are being misused by the libertines (above on Acts 20.29, Rom. 3.8). He complains not of misunderstanding, but of deliberate misrepresentation, and he does not write as if the persons concerned were his former disciples. But if the misinterpretation was deliberate it was probably in the interest of some already existing position.

(Paul was also misrepresented, by the opposite side, as preaching circumcision, Gal. 5.1 1.) Moreover, there is another figure to whom all the early and widespread branches of the libertine tradition can plausibly be traced, since attacks on his liber tine teaching and practice are prominent in the reports of his work. He broke the sabbath, he neglected the purity rules, he refused to fast, made friends with publicans and sinners, and was known as a gluttonous man and a winebibber. He not only taught his disciples that the law had come to an end with the Baptist and that the least of those in the kingdom was greater than the Baptist, but he also administered a baptism — "the mystery of the kingdom of God" — by which he enabled some of his disciples, by union with himself, to enter the kingdom and to enjoy his own freedom from the law.

THE BACKGROUND

Therefore, in our picture of pre-Pauline Christianity, alongside the legalistic interpretation of the religion we must set the libertine. The legalistic interpretation went back to the (principally Pharisaic?) converts of the Jerusalem church, and appealed to the tradition of Jesus' exoteric teaching. The libertine interpretation went back to Jesus himself and preserved elements of his esoteric teaching. It was dominant in the Jerusalem church in the earliest days, but lost its hold as the small group of Jesus' original, initiated disciples was outnumbered by the new converts under the leadership of Jesus' brother James. We see the leader of the old disciples

— Peter — challenged by the legalists in Acts 1 1 ; he disappears from the city in Acts 12; in Acts 15 is he back again — but is Acts 15 historical? Anyhow, James presides at the meeting. In Acts 21 only James remains. When Paul visits the city he is persuaded to make a public pretense of keeping the law. Paul, by virtue of his donations, was doubtless a highly valued member of the Church, moreover he could recall and might report the traditions of the original apostles, so James was willing to make adjustments — for Paul. As for his general policy, however, the mere appear ance of his representatives in Antioch was enough to frighten Peter out of eating with gentiles (Gal. 2.12). At least in Jerusalem the libertine tradition would be a secret doctrine.

D. The " loss " of all writings of Jesus'" immediate followers This leads to a fourth consequence of Jesus' baptismal practice, this time an indirect consequence: the "loss" of all writings from Jesus and his immediate dis ciples. That they were all illiterate is improbable. They founded a sect and the survival of the sect makes the disappearance of all works of its founders a noteworthy problem

(Brandon, Fall qf). Having seen how Jesus' immediate disciples disappeared from the Jerusalem church, we may suppose that their writings went with them. The libertine consequences of Jesus' baptismal practice probably helped James and his party to get rid of both the apostles and their writings.

James's works, if any, have also disappeared and we have nothing from the legalist

Jerusalem church of a.d. 40-66, but these facts are largely explained by the great revolts of 66-73 and 1 32-1 35 in Palestine and of 1 1 5-1 17 in Egypt, Cyrene, Cyprus,

Syria, and Mesopotamia, which must have ruined most of legalistic Christianity (so Brandon, Fall). Asia Minor was the only area in the Near East where a large

Jewish population survived relatively undisturbed, and Asia Minor had been the scene of Paul's most successful work. This probably was a major cause of the predominance of Pauline material both in the NT canon and in the entire corpus of early Christian material.

However, besides this external cause it is plausible to attribute the success of Pauline

Christianity to its practical nature. In the conflict between the libertine and the

legalist interpretations, Paul represents the safe and sane and socially acceptable compromise. He accepts the realized eschatology of the libertines (Rom. 14.17;

Col. 1. 13) but rejects those who make it an excuse for refusing to work (II I hess.

CLEMENT OF ALEXANDRIA

2. iff, 3·6ίΤ). He also accepts the expectant eschatology congenial to the legal inter pretation (I Thess. 5. iff; II Thess. 1; etc.) but he tacitly drops the nationalistic and political side of it which were to involve the legalists in the military disasters of the rest of Judaism (Brandon, Fall i67ff, esp. 180). He also rejects the revolutionary implications of Christianity in the political and social fields. Jesus' record on taxes is equivocal, at best (above, p. 249); Paul is explicit: they are to be paid (Rom.

13.6). Subjection to the existing political authorities is not only a wise policy, but a moral obligation (Rom. 13.5). Wives, obey your husbands; slaves, your masters

(Col. 3.18,22). In matters of personal morality he reverences the Mosaic Law (which had immense prestige) — it is holy, just, and spiritual (Rom. 7.12,14) — but he rejects the observance of it and especially of circumcision and the food laws, which did so much to hinder the growth of Judaism and to isolate the Jews as a peculiar people — a dangerous social position. Rejection of the rules of the law leaves him free, and the claim of freedom was highly valued in a world much influenced by the Stoics and the Cynics. But Paul's liberty is guided by the spirit, and by common sense (I Cor. 11.14, "nature itself"); and the spirit (very much like common sense), teaches industry, thrift, obedience, good citizenship, moral propriety, and all those other requirements which must be observed if the individual is to succeed in society, or if a small society — in this case, a local church — is to succeed in the larger world.

What early Christianity was like in the areas from which no material has been preserved can be inferred from the things Paul opposed in his letters, and this explains why no material was preserved.

E. Second-century silence about PauVs doctrine of baptism Finally, we may consider as a more remote consequence of Jesus' baptismal practice the disappearance of the baptismal teaching of Paul himself from the works of the second-century Christian writers of the party ultimately triumphant ("ortho dox"). As Benoit, Bapteme , shows, the orthodox references to baptism throughout this period consistently stress the remission of actual sim (not original sin), and the gift of the spirit as an earnest of future life (not a present spiritual power and source of moral behavior, but a vague deposit to be protected by moral behavior). Notions of new birth, illumination, and exorcism occur but are comparatively unimportant.

The essential Pauline theme of death and resurrection with Christ is totally absent.

Benoit insists (p. 228) that this is not due to misunderstanding of Paul; there are no reminiscences of his works. Hence Benoit concludes that orthodox second-century

Christianity came largely from non-Pauline sources, and he inclines to agree with Bauer ( Rechtglaubigkeit 221, 227, 230) that Paul was most venerated in heretical circles (p. 229).

But this will not do, because most of the evidence for the early history of Chris

tianity is connected with Asia Minor: Lk.-Acts , Jn., and the Johannine epistles, the Pauline epistles , the pseudo -Pauline epistles (including Hebrews), I Peter , James, the Apocalypse, the Ignatian epistles, Polycarp, Papias, the epistle of Pliny, the Martyrdom

THE BACKGROUND

of Polycarp, Marcion, Justin's Dialogue (laid in Ephesus), Melito, Polycrates, Avircius

Marcellus and his anti-Montanist correspondent (Eus., HE V. i6f), the presbyters of Irenaeus ,

Irenaeus himself — all are examples which immediately come to mind, and the list could be extended. All of the writers and works italicized in this list show clear

Pauline influence. More than that, the predominance of this material in the corpus of early Christian evidence and the predominance of Pauline material in the NT are mutually confirmatory evidence. As remarked above, the NT is a product of selection — it shows which, of the many first-century Christian documents, the party that became predominant in the late second century chose to keep. So, to a lesser degree, is and does the remainder of the preserved corpus of early Christian material

(Goguel, review of Bauer 164). The material chosen by the triumphant party was predominantly Pauline.

Consequently the disappearance of Paul's theology of baptism from second-century

Christianity cannot be explained as Benoit proposes— by ignorance. It must have been deliberate. But if Paul represented safe, sane, and successful Christianity, and if it is for this reason that the Pauline tradition dominates the canon of the NT and the corpus of early Christian literature, why did second-century Christianity drop his doctrine of baptism ?

The answer, I think, must be sought along the lines indicated above. Paul's doctrine had been a safe compromise by comparison with the competing extremes of his day, but for baptism he had substantially preserved the teaching of Jesus; and as the potentialities of this teaching became apparent, it also became apparent that Paul had not been safe enough. (A similar explanation has now — 1965 — been given by Barrett, Things Sacrificed .) The Jewish revolts had ruined forever the legalistic interpretation of Christianity, but the libertine interpretation was flourishing and it found in Paul's doctrine of baptism everything needed to justify its inclinations: identification with Jesus, possession by his spirit, ascent to the heavens in his ascension, and consequent liberation from the law. Accordingly, the gnostic teachers, who now emerged from the libertine tradition, made great use of Paul (Harnack,

Quellenkritik 27, 29, etc.); and Grant, Gnosticism 149, finds it difficult to decide whether his teaching "is more adequately reinterpreted by these gnostic teachers, or by such second-century 'orthodox' teachers as the apologists and Irenaeus."

Buri, Clemens, has shown in detail how drastically Paul's concept of freedom was remodeled by the orthodox in the interests of social acceptability. The process can be seen in its simplest form in Acts, where Paul's career is reported, his theology omitted. That the letters had been esoteric documents to begin with made matters easier.

XII. Conclusion: the historical value of the new text

It has now been argued that the interpretation of "the mystery ol the kingdom of God" as a baptism with Jesus' spirit, effecting entrance of the kingdom, will

CLEMENT OF ALEXANDRIA

explain (A) the coming of the spirit in the early churches; (B) their adoption and development of baptism; (C) the primitive and widespread libertine tradition within Christianity and the consequent persecution of the early churches, as well as their internal political struggles; (D) the "loss" of all writings of Jesus' immediate follow ers; and (E) the abandonment of Paul's doctrine of baptism by the predominantly

Pauline "orthodox" churches of the second century. In the course of the argument it has become evident that this interpretation would also explain the extent and importance of secret doctrine in the early churches, and would indicate the content of the most important secrets. Will the reader please offer another explanation for all these problems?

While awaiting another explanation, we shall consider the history of Christianity, in the first and second centuries, as a vector of the conflict between the libertine tradition and its opponents — in the first period the legalists in Jerusalem and Paul in the Aegean area; later the developing institutionalism of the established churches and the ascetic movement probably to be explained by influence from Greek philo-: sophy. The situation in the later period has been given classic description by Bauer,

Rechtglaubigkeit. The state of affairs he discovers in the early second century is inexplicable unless we suppose a massive libertine development in the first century; and such a development at so early a date must be taken as indicative of something in the origins of the movement.

It is against this background that the problem of the origin of the Gospels must be seen; and against this background the new text would seem to be of the highest historical value.

XIII. Carpocrates Of the several parties in this background, the libertine has been the least studied.

The Carpocratian sect was one of its most extreme representatives, and is directly involved in the history of the longer text of Mk. Therefore it seems worthwhile to review here the evidence concerning the Carpocratians. The texts are collected in Appendix B.

A. Names Καρποκράτης with K is an extremely rare name, but a woman, Καρποκρ[άτη( ?)], appears in P.Oxy. 1473 line 24 (a.d. 201), and Κα[ρ7τοκρά]της as the name of the deity appears in the hymn to Isis, on Ios (3 c. b.c., SIG 1267). Άρποκράτης or Άρποχράτης with the rough breathing is somewhat commoner (P. Teb. I, no. 83;

Viereck, Ostraka no. 365; etc.). Epiphanes as a personal name is not uncommon ( P . Erlangen , no. 60 line 9; SEG XIV. 685. 3; P. Teb. III.1.811 line 23; etc.).

THE BACKGROUND

B. Date As to the date of Carpocrates we have three types of evidence. First, explicit statements: Eus. (HE IV. 7.9) says Irenaeus said Carpocrates was a contemporary of Saturninus and Basilides. Irenaeus does not say so in his preserved works, and the probability is that Eusebius was arguing from the fact that in Irenaeus' account of heretics Carpocrates immediately follows Saturninus and Basilides (so Usener,

Weihnachtsfest 116). But Irenaeus did not treat the heretics in chronological order: for example, he puts Cerinthus after Carpocrates and Carpocrates after Basilides, although he himself tells of Polycarp's report of a near-encounter between Cerinthus and John the apostle (Harvey, III. 3. 4 = Stieren, III. 3. 4) whereas Basilides, com monly placed in the third generation of Simon Magus' line, probably flourished in the time of Hadrian (117-138 — Clement III.75.13ff). Theodoret (Compendium 1. 5 end) says flatly that Carpocrates and Epiphanes flourished in the reign of Hadrian; he may have derived this information by inference from Eusebius, or he may have had some other source. Michael Syrus dates Carpocrates in the time of Bar Kosibah and "the prince Antinous," Hadrian's favorite, but Michael was certainly using

Eusebius.

In the second place there are statements about the history of the sect. The often repeated statements that Cerinthus was Carpocrates' successor or that Carpocrates followed Basilides merely reflect the order of names in Iranaeus (or his source), which, as we have seen, is not chronological. Epiphanius declares that Epiphanes,

Carpocrates' son, who died at the age of 17 or 18, joined the sect of Secundus, the disciple of Valentinus. But Epiphanius is an extremely unreliable writer, and in this instance his statement is probably based on a misunderstanding of Irenaeus' reference to an h ηφανής διδάσκαλο? of the Valentinians. (So Holl in GCS on Epi phanius, Panarion XXXII. 3, referring to Irenaeus — Harvey, 1. 5. 2 = Stieren, I.11.3.)

Far more important is Irenaeus' statement that Marcellina, a member of Carpocrates' sect, came to Rome under Anicetus (ca. 150-165), and had great success there

(Harvey, 1. 20. 4 = Stieren, 1. 25. 6). The branch of the sect which followed Marcellina differed so greatly from the original that Celsus, about 175, thought them different sects (Origen, Contra Celsum V.62). Accordingly, a date for Carpocrates no later than the generation preceding Marcellina (i.e., ca. 125) is indicated. (Praedestinatus' statement that the apostle Barnabas condemned the Carpocratians in Cyprus is probably worthless as evidence of an early date.)

Finally, there is the evidence from the lists of heretics. A conspectus of the older lists is given at the beginning of Appendix B, where the different symbols, marking the different groups of heretics which commonly are listed together and in the same order, show clearly the sources (whether documents or oral traditions is immaterial) from the which present lists have been put together. The core was a list ol Simon

Magus and his successors : Menander, Basilides, Satornilus. This is understandable : at the beginning, as Justin makes clear, Simon had been the great competitor (First

Apology 56). Next comes a group composed of Carpocrates (with or without Mar cellina), Cerinthus, and Ebion. (To this Epiphanius adds some Jewish-Christian

CLEMENT OF ALEXANDRIA

heresies.) Around these two central groups (Simonians and Carpocratians) different groups have been added in different places by different writers. The group of Samar itan and Jewish heresies, when added, is of course put first; but it is certainly an accretion, absent from Justin, Celsus, Irenaeus, and Hippolytus. The Valentinian group is also secondary, as shown by the variation of its position. It was added in front of the Simonians by Irenaeus and Hippolytus, for whom Valentinus was the major, present competitor, but was moved behind the Carpocratian group by pseudo-Tertullian, Epiphanius, and Filastrius, who wrote when the Valentinian peril was past. The Ophites and their ilk are added sometimes at the end (Irenaeus), sometimes at the beginning (Hippolytus, Filastrius), sometimes at one or another place in the middle (pseudo-Tertullian, Epiphanius). Finally, the schools associated with Marcion and Tatian are always placed at or near the end, in accord with the general opinion as to their relative date.

The surprising thing about these lists, for our purposes, is not only the early date they indicate for Carpocrates, but also their persistent classification of him with the Jewish-Christian sects of Cerinthus (on whom Dionysius of Alexandria, in Eus., HE

VII. 25) and the Ebionites. This is odd, since the Platonic traits in Carpocratianism have often been remarked — recently, e.g., by Festugiere, Revelation 1 1 1. 2 5 — and were noticed as early as Irenaeus (Harvey, II.5of = Stieren II.33f) and Clement

(Π .200.i6ff). A further peculiarity is the fact that this tradition puts Carpocrates at the head of the Jewish-Christian group, although Cerinthus and the Ebionites were presumably earlier. We must look for explanations of these facts, and shall do so by examining in sequence the testimonia collected in Appendix B — with first a word about Jude and II Pet.

C. Testimonia Clement (II.200.24ff) says that Jude in his epistle "spoke prophetically" of the Carpocratians and similar heretics; and in the letter, 1. 3-6, he identifies the erring stars of Jude 13 as Carpocratians and says they claim to know the deep things of Satan — a claim which in Apoc. 2.24 is attributed to the followers of Jezebel of Thya tira. Its transference to the Carpocratians is rhetorical. What justification there was for the rhetoric is unknown. The notion that Jude refers to the sect may be correct.

Does II Pet. also refer to them and was the reference intended by the common source of Jude and II Pet.? These questions are insoluble, therefore the details in Jude and II Pet. cannot safely be used as evidence of Carpocratian teaching although they may be so. But in order to explain Clement's interpretation, we must suppose the general picture given by Jude bore some similarity to Carpocratianism; this is important since the source of Jude and II Pet. may have had connections with early Syrian Christianity.

Now for the texts printed in Appendix B:

Heracleon (ca. a.d. 160) quoted by Clement in Eclogae propheticae 25.1 (III. i43.22ff) spoke of certain Christians who branded their ears and who justified the practice

THE BACKGROUND

by allegorization of Lk. 3.17 || Mt. 3.12. That he referred to the Carpocratians is not certain, but it is made likely by the context of the following reference in Gelsus.

Celsus ' report (ca. 1 75) must be disentangled from the comments of Origen, who pretends ( Contra Celsum V.64) that only by misunderstanding I Tim. 4.1-3, iv ύστεροίς καιροϊς άποστησονταί Tires' της ττί στεως, προσεχοντες ττνευμασί πλάνοίς καί διδασκαλίας δαιμόνιων, εν νποκρίσεί φευύολόγων, κεκαυστηριασμενων την οίκείαν συνεί'ύησιν, κωλυόντων γαμεΐν, άπεχεσθαι βρωμάτων, and also misunderstanding those who used these apostolic words against heretics, could Gelsus have been led to believe that some Christians would call others such horrid things as Κίρκας . . . κύκηθρα αίμνλα . . . ακοής καυστηρια . . . αινίγματα . . . Σειρήνας εξορχουμενας . . . σοφιστρίας, κατασφραγιζομενας τα ώτα, καί άποσυοκεφαλονσας τούς ττειθομενους.

Here Origen is misrepresenting Christian party polemics. There is no reason to doubt that Celsus got these phrases from a Christian tract which must have antedated his work. Heretics are compared to sirens by Hippolytus, Philosophumena VII. 13.

The reference to branding the ears, together with libertinism (Circes . . . making pigs of their followers), suggests that the polemic was directed against the Carpocra tians. Branding is reported of some branches of the sect by Irenaeus (Harvey 1. 20. 4 =

Stieren 1. 25. 6); and became notorious. Besides the numerous references in the later testimonia, see Minucius Felix 9: Eruenda prorsus haec et exsecranda consensio.

Occultis se notis et insignibus noscunt et amant mutuo, paene antequam noverint; passim inter eos velut quaedam libidinum religio miscetur , ac se promisce appellant fratres et sorores , ut etiam non insolens stuprum intercessione sacri nominis fiat incestum. On the other hand, it is unlikely that I Tim. 4. iff refers to the Carpocratians, since, as the context shows, the opponents observe food taboos. The "branding of their consciences" is probably a sneer at some initiatory branding; if so the practice was not limited to the Carpo cratians. Dolger, Sphragis 77Γ, found evidence suggesting similar practice, in later sects, interpreted as a baptism of fire (to fulfill Mt. 3.1 1, Lk. 3.16, contrast Heracleon). Dolger thought this a second baptism {ibid., 75ff). Jerome, Luciferianos 23 end, would then be mistaken — not an unlikely supposition: in the same sentence he mistook Ebion for a personal name: see Vallarsi's pathetic note, ad loc.

On the other hand, Dolger was probably mistaken in supposing the rite came from the Greek world; he overlooked the OT provision for piercing the ear of a slave as a sign of perpetual servitude (Dt. 1 5 . 1 6f )- — in this instance of Christ — because he loves his master and finds life good in his master's house. This practice was still in force in rabbinic times [T. Baba ' Kama ' VII. 5; Sifire Devarim 121; etc.). Moreover,

Bousset, Antichrist 25 and 20off, has shown that the expectation of sealing is probably part of the pre-Christian apocalyptic tradition, cf. the rabbinic tradition that Jesus

(? = NltfO ]l) came back from Egypt with magical names or signs tattooed on his skin (J. Shabbat XII.4(i3d), B. Shabbat 104b). Such marking of the body as part of initiation into societies of magicians is probably prehistoric.

Origen's quotation from Celsus gives a list of the sects to which Celsus referred ca. 1 75 to prove the varieties and self-contradictions of Christianity — no doubt the sects best known in his environment. Origen says he called "the Church {sc.

CLEMENT OF ALEXANDRIA

Origen's party) οι από του πλήθους ; these are they who say their god is the same as the god of the Jews (V.61). Besides these there are some unnamed dualists, then the Valentinians, Gnostics, Ebionites, Sibyllists, Simonians from Helen, Marcellians from Marcellina, Harpocratians from Salome, others from Mariamme (Ophites:

Hippolytus, Philosophumena V.7.1; X.9.3), others from Martha, and finally the Marcionites. Notice the number of libertine sects (Gnostics, Simonians, Marcellians,

Harpocratians, Ophites, and possibly others), the number "from" women, and the importance of the Carpocratians, who account for two of the eleven. The impor tance of women in the practice of magic appears already in Enoch 7. iff: it was to women that the fallen angels first taught the black arts. In rabbinic literature it is a commonplace: even the purest of women practices magic (J. Kiddushin IV. 11

[66c end]); a woman being punished in public is automatically supposed to have practiced magic or committed adultery (Sifre Devarim 26 || Wayyikra Rabbah 31.4); see further B. Berakot 53a end; B. 'Erubin 64b end; Sotah IX. 13; etc. Hence the importance of women in libertine Christianity and in the circle of Jesus' disciples

(Lk. 8. iff, etc.; above, commentary on II. 23) is probably significant. Salome, in particular, indicates some connection between the Carpocratians and the Gospels of Mark, Thomas , and the Egyptians (see above, pp. 1 8gfF) .

Origen's statement that he has never met any of "these" is of uncertain reference

— and veracity. However, the libertine wing of Christianity does seem to have diminished sharply in the early third century, perhaps because of police action. In Celsus' comparison with the worshipers of Antinoiis, where does the similarity stop and the polemic begin? The main point of similarity is immoral behavior; this fits later Christian reports of the Carpocratians; so do the repeated references to ear-marking in Celsus' source, τους παραχαράττοντας τα Χριστιανισμού , if it came from Celsus' source, may be echoed in the letter of Clement 1. 14, but is too common to prove relationship.

Hegesippus , ca. 180, in Eus. HE IV. 22.46?, derives all Christian heresies from Samaritan and Jewish-Christian sects, and therefore is not a strong witness for this derivation. The order agrees with Justin's (

Dialogue 35.6) save for the presence of the Carpocratians; unfortunately, it is probably not chronological.

Irenaeus , ca. 185, provides the only account of Carpocratian theology which can pretend to reliability. Clement's reports are mostly limited to questions of ethics, and the later writers have little, if any, independent knowledge of the sect. The theology described by Irenaeus has many apparently Platonic traits: the creation of the world by inferior beings, salvation for the pure soul able to remember what it saw when moving in the celestial sphere, transmigration as a penalty for souls which have not attained perfection. It also uses the Stoic notion that physical actions are indifferent by nature, good or bad only by opinion.

But in the first place, some of the apparently Platonic traits may not have come from Platonism. For instance, the creation of the world by inferior beings may be one consequence of that awareness of the importance of intermediaries which appears

THE BACKGROUND

in every field of thought during the imperial period and is probably a reflection of the imperial administration: it appears where Platonism does not — for instance, in Paul's notion that God gave the law through intermediaries (Gal. 3.19).

In the second place, the Platonic traits are superficial and explanatory; beneath them is the structure found in Col. 2.8-3.45 and in the Typhon initiation quoted above, on page 221 : the soul's escape from the rulers of this world by union with a spirit which gives it magical powers (and, in Col. and Carpocrates, enables it to ascend into the heavens). Platonism helped make this sort of thought respectable, but the practice was magical and, in Carpocrates' case, Christian. That libertine

Christianity was Platonized in Alexandria is understandable; the Platonic and Stoic traits in the system of Carpocrates are no more proof of Platonic or Stoic origin than are those in the systems of Clement and Philo.

Turning to details: the report that Jesus was begotten by Joseph shows the Carpocratians holding to an early tradition the majority abandoned.20 Hilgenfeld, Euan geliorum 18 and 36, conjectures that they used Mt. with the Sinaitic Syriac reading at 1. 1 6 (cf. Epiphanius, Panarion XXX. 14). The report of Jesus' training and escape is given by Irenaeus from two, mutually explanatory, sources printed below in parallel columns. The words added in Greek are from Hippolytus ( Philosophumena

VII. 32) who evidently read them in his text of Irenaeus though they are not in the present Latin translation. Epiphanius, Panarion XXVII. 2. 2, paraphrased the same passage of Irenaeus; his paraphrase shows that he also found these words in the text.

B A Jesum autem . . .

Jesu autem animam distasse a reliquis <( hominibus ) in Judaeorum consuetudine nutritam γεγονότα όικαιότερον των λοιπών secundum id quod anima ejus firma et munda cum esset commemorata fuerit quae visa essent sibi contempsisse eos (cf. the following sen tence : contemnere mundi . . . archontas)

. . . ingenito Deo et propter hoc ab eo missam esse ei virtutem et propter hoc virtutes accepisse uti mundi fabricatores efiugere posset et per omnes transgressa et in omnibus per quas evacuavit quae fuerunt in poenas liberata ascenderet ad eum passiones 20. Another primitive element may be their denial of the physical resurrection. Richardson suggests that the diversity and disagreements of the resurrection stories in the Gospels (striking by contrast with the close agreements of the preceding passion stories) indicate that the stories ol a bodily rcsuirection are accretions to the original preaching, which declared only, "he is alive again sc. as a life-giving spirit (Acts 25.19; I Cor. 15.45).

CLEMENT OF ALEXANDRIA

That Jesus was brought up and excelled in the observance of Jewish law would explain why he appears in the Gospels as a legal authority — a problem which has no other clear answer. It would ht also with Mk.'s location of the dispute stories, and Jn.'s of the cleansing of the temple, early in Jesus' career. The Platonizing explanations in text A ( anima . . . munda . . . commemorata , etc.) tell how Jesus was enabled to see through the cosmic powers (equated in text B, as in Col. 2.16-23, with Jewish observances). The reward of this philosophic penetration is the descent of the holy spirit— the virtus of text A (= δόνα μις, which leaves Jesus only at cruci fixion, Gospel of Peter 19 || θεός, Mk. 15.34; Χριστός, Cerinthus in Irenaeus [Harvey,

1. 21 end — Stieren 1 . 26. 1 ] ; Simon Magus also was or had a δυναμις, Acts 8.10; further, Vaganay, Evangile 108). The spirit makes it possible for Jesus to overcome the cosmic powers (which text B now identifies with the passions) and so to ascend to God. Per omnes transgressa et in omnibus liberata, the means of victory, reflects the ascension theory discussed above, pages 238ff. Each cosmic sphere had its proper demonic ruler, physical element, and psychological passion. Consequently, ascent involved not only movement through space and victory over the demons, but also change in physical nature and victory over the passions. But in order to overcome a passion, one must experience it. The consequences of this belief appear in Carpocrates' theology and in that of the contemporary ( ?) Epistle to the Hebrews, 2.9ff and q.iqff.

( χωρίς αμαρτίας, tagged to the end of verse 15 as a safeguard, suggests the author knew libertine developments of the theory and was trying to avoid them.) Another common belief was that the spirits of άωροι — persons who died before their time, especially before marriage — became earthbound demons subject to magicians and to the chthonic gods. Hence Carpocrates' teaching that complete experience of this life is necessary for liberation from it. The Carpocratians' support for this theory

— their interpretation of Mt. 5.25 || Lk. 12.58 ("agree with your creditor quickly . . .")

— seems to have been current in Paul's time (above, p. 259). On the antiquity of other demonological exegeses handed down in secret tradition, see Danielou, Tradi tions 200.

Apart from this basic structure of Carpocrates' theology, the most important things for our purpose in Irenaeus are his reference to Carpocratian texts containing secret teachings of Jesus — this accords with the report in Clement's letter — and the indirect evidence he provides for the influence of the Carpocratian sect in his environ ment. He says (Harvey, 1. 26. 2 = Stieren 1. 28. 2) that Basilides and Carpocrates provided the starting points of all the libertine sects. Again (Harvey II. 48-53 =

Stieren II. 3 1-33), after explaining that his refutation of the Valentinians refutes by implication all other heretics, he deals first with elements common to the Carpo cratians and other sects (creation by inferior angels, possession of supernatural powers, and ability to do miracles) and next with claims primarily Carpocratian: that one must experience everything, that they have souls from the same sphere as Jesus' soul, that Jesus did his miracles by magical delusions ( φαντασιώδους ), that souls are punished by transmigration. This extended attack on the Carpocratians shows that Irenaeus thought them the most important of the libertine sects against which he had to argue.

THE BACKGROUND

Clement (ca. 190) was of the same opinion. When he came, in Stromateis III, to deal with the libertine heresies, he took the Garpocratians first and treated them at far greater length than any of the others. Yet— and this is typical of Clement — he gives no systematic account of the Garpocratian position. He represents Epiphanes,

Carpocrates' son, as the source of both "the monadic gnosis" and "the Garpocratian sect' ' (II. I97.26ff). The letter shows that the second part of this is polemic: the sect came also from Carpocrates and had behind it the longer text of Mk. (above, p. 84). And the first part of Clement's statement is no more reliable. Chadwick has pointed out ( Alexandrian Christianity 27) that the monadic gnosis probably had behind it "an apocryphal work . . . the mother of their licentiousness"

(II.209.17ff).

From Epiphanes' On Justice Clement quotes a considerable passage representing the physical world as the work of a just — that is, egalitarian — god, interpreting Mt.

5.45 as evidence of this, and consequently describing Ex. 20.17 ("thou shalt not covet") as "ridiculous" and the law in general as an occasion of sin because it runs counter to divinely established communism. (Here Clement saw a misunder standing of Rom 7.7.) This might be thought to contradict Irenaeus' statement that Carpocrates said the world was made by inferior angels; but Epiphanes, being a Platonist, probably held that daimones had created the world according to a divinely given form; he could therefore speak without self-contradiction of divinely ordained patterns in the demonic creation.

Next Clement reports the orgiastic agapai of the Carpocratians. Jude 12 (cf.

II Pet. 2.13) declares the heretics it attacks particularly active in agapai. Clement's report is suspiciously similar to those that contemporary pagans were spreading about all Christians (cf. Tertullian, Apology VII. 1), but he adds that in such agapai they seek communion (II.200.12) and later, when discussing the libertine sects generally, he protests the use of κοινωνία μυστική for sexual connection and he lets slip ταύτην οΐονται εις την βασιλείαν αυτούς άνάγειν του θεοϋ (ΙΙ.2θ8. ioff; 24f)> showing that the primitive functional connection of the sacraments with the kingdom persisted in the sect. (Here, by the way, we have another reference to the Carpocratians as ot παραχαράσσοντες την αλήθειαν, as in the letter, 1 . 1 4. )

Besides their ritual copulation in the agapai, Clement says, they also ask of any women they wish performance of the "divine law" (II.200.13ff), which he suggests they derived from a "misunderstanding" of Plato. But eight pages later, in a passage where the Carpocratians are not named (the same passage which contains the information about κοινωνία quoted above), the "divine law" turns out to be Jesus' saying "give to him who asketh of thee," which the wicked quote from Scripture

(γεγραττται). Clement's attribution to Plato was a smoke screen.

After remarks on several particular sects, Clement proceeds to more general discussion and for the purpose divides all heresies into two groups, the ascetic and the libertine. Much of what he says about the libertines probably refers to the Carpocratians, whom he had taken as the first major example of the type. It is

probably they who appeal to the sayings "the Son of Man is Lord of the Sabbath (II. 214. 17), "all things are permissible" {ibid., 19), and "you were called to liberty"

CLEMENT OF ALEXANDRIA

(ibid., 28f). (The Pauline contexts of the two latter sayings are notably favorable to Clement's purpose and unfavorable to libertinism; perhaps the sayings were agrapha which Paul was explaining away and which the libertines quoted as agrapha, not as Pauline.) The attack on those who maintain επιθυμία χαριστεον καί τον επονείδιστου βί ον αδιάφορου ηγητεον suggests the whole passage is aimed at the Carpocratians. This fits Irenaeus (Harvey 1. 20. 3 end = Stieren 1. 25. 5) : the Carpocratian report of Jesus' secret teaching was per fidem enim et caritatem salvari; reliqua vero, indifferentia.

In II.221.6ff Clement mentions the Carpocratians and attacks their interpretation of "give to him who asketh." This develops into a defense of wealth as the prerequisite of charity, and the defense makes it seem that Clement's Carpocratians practiced economic, as well as sexual, community of goods. They evidently appealed to Mt.

1 9.2 if: "If you wish to be perfect, sell what you have and give to the poor" — we met the Markan parallel in the baptismal pericope. (Above, pp. 172b This may explain why η γυναίκα has disappeared from Mk. 10.29; cf. Theophylact on 10.30, άρα οΰν καί γυναίκας εκατονταπλασίονας ληφεται) ναι, καν ό κατάρατος ' Ιουλιανός εκωμώδει τούτο. Julian's joke may have referred to stories derived from Carpocratian practice.

Cf. also Origen, Exhortatio ad martyrium 16.) In II.224. i off we again have an attack on those who advocate αδιαφορίαν. This time they appeal to Rom. 6.14 and do not believe in the last judgment. In the concluding section of book III (II.244.21ff), the community of wives and economic communism recur, together with some new traits : the Carpocratians' notion of την διαθήκην την καινήν, their use of baptism, and their importance as a cause of the bad reputation of Christians in general. This section has a number of traits also found in the letter's attack on Carpocratians (πάντα καθαρά τοίς καθαροϊς, εις το εξώτερον σκότος, etc.) . On all the material in Stromateis III see now Bolgiani, Polemica. Further pas sages in Clement's works may also refer to the Carpocratians; for instance, III.143.12ff:

Οτε χοϊκοί ήμεν ; Καίσαοος ήμεν. Καΐσαρ δε εστιν 6 πρόσκαιρος αρχών, ου καί είκών η χοϊκή 6 παλαιός άνθρωπος, εις δν επαλι νδρόμησεν. τουτω ούν τα χοϊκά άποδοτεον, ά

" πεφορεκαμεν εν τη είκόνι του χοϊκού ," καί "τά του θεού τώ θεώ· " έκαστον γάρ των παθών ώσπερ γράμμα καί χάραγμα ημϊν καί σημεΐον. άλλο χάραγμα νυν ό κύριος ημΐν καί άλλα ονόματα καί γράμματα ενσημαίνεται, πίστιν αντί απιστίας, καί τά έξης, ούτως από των υλικών επί τά πνευματικά μεταγόμεθα " φορεσαντες την εικόνα του επ ουράν ίου."

This might be interpreted to fit Carpocratian doctrine; and it immediately precedes, in the Eclogae propheticae, Clement's quotation from Heracleon referring to the sectarians who branded their ears. Moreover, iqff explains the equivalence of demons and passions in Irenaeus' accounts of Carpocratian teaching, quoted above. As John of Damascus would have said, the demons are one in form with the passions they have given us. Like the orthodox, the Carpocratians were groping for philosophical explanations of the gospel sayings — and of the magical practices — which had come down to them from their Palestinian origin.

After Clement the material in Appendix B affords almost no reliable information about the Carpocratians.

THE BACKGROUND

Tertullian exploits Irenaeus, Hippolytus abbreviates him, Epiphanius inflates him.

Occasionally Hippolytus and Epiphanius make it possible to improve the text used by the Latin translation of Irenaeus (the most important example was given above.) Epiphanius furnishes additional material, but because furnished by him it is so unreliable that it does not add to our knowledge. Thus, when he makes the Carpocratian god όρχη, άγνωστος, ακατονόμαστος and declares that the angels revolted from him before creating the world, he may be importing traits from other gnostic systems ; when he specifies the Carpocratians' wicked actions, he is probably indulging his libidinous imagination; and when he credits them with two kinds of images of Jesus — a secret set modeled on Pilate's and another used with the images of philosophers — he is probably confused. That, when Carpocratians are known, gentiles will have nothing to do with Christians, is probably his own improvement of the scandal story. That Carpocratians come near Christian groups only to snare souls may be his excuse for the fact that none are to be seen (Origen, over a century earlier, seemed to say he had never been able to meet one, and he may have been telling the truth). Epiphanius' connection of Epiphanes with Secundus is merely a muddle, and his statement that the Origenians resemble the sect of Epiphanes means only that they had filthy practices. Perhaps his report that the Carpocratians used Mt., with the genealogy, to prove the human birth of Jesus, is reliable, but even about this there is no being sure.

From the minor authors in Appendix B there is even less to glean.

PseudoTertullian'1 s notion that Jesus' ascent to the heavens, according to the Carpo cratians, took place after his death is a new specification and, from the evidence we have seen, presumably a misunderstanding: Jesus' role in Carpocratian theology as revelator requires that he have already ascended and seen what he reveals — the more so since the Carpocratians held to the historical fact of his human nature; contrast John i . iff.

Eusebius we have already spoken of. He adapts Irenaeus' information to his own apologetic interests, using the Carpocratians to explain the scandalous charges — including incest — which have been brought against the Christians.

Athanasius reflects Irenaeus.

Marcellus of Ancyra gives a list of sects in unusual sequence, but his originality results from confusion, not knowledge.

Gregory's remark shows that the sect had disappeared quite a while before the rise of the Arian controversy, but was remembered as one of the six or seven main

↑ Back to top