Codex 2400 and Its Miniatures 9781463233921

Citation preview

Codex 2400 and Its Miniatures

Analecta Gorgiana

990 Series Editor George Anton Kiraz

Analecta Gorgiana is a collection of long essays and short monographs which are consistently cited by modern scholars but previously difficult to find because of their original appearance in obscure publications. Carefully selected by a team of scholars based on their relevance to modern scholarship, these essays can now be fully utilized by scholars and proudly owned by libraries.

Codex 2400 and Its Miniatures

Harold R. Willoughby

2013

Gorgias Press LLC, 954 River Road, Piscataway, NJ, 08854, USA www.gorgiaspress.com G&C Kiraz is an imprint of Gorgias Press LLC Copyright © 2013 by Gorgias Press LLC Originally published in 1933 All rights reserved under International and Pan-American Copyright Conventions. No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, scanning or otherwise without the prior written permission of Gorgias Press LLC. 2013

O

ISBN 978-1-61143-943-4

ISSN 1935-6854

Reprinted from the 1933 New York edition.

Printed in the United States of America

C O D E X 2 4 0 0 AND ITS MINIATURES By H A R O L D

R. W i L L O U G H B Y 1

T

H E Rockefeller McCormick m a n u s c r i p t — t o cite at once the familiar designation for C o d e x 2400—is a complete Byzantine N e w Testament, dating from the thirteenth century, with nearly one hundred brilliantly illuminated pages. 2 T r u e , it does not include the A p o c a l y p s e of John. T h i s omission, however, is due to no defect in the manuscript itself, but rather to the irregularity of the Eastern canon of Christian scripture. S o tardy was the book of Revelation in winning recognition for itself in the Greek East, that manuscripts of the G r e e k N e w Testament which conclude with the A p o c a l y p s e are comparatively rare. G r e g o r y lists only fifty-three N e w Testaments that include the A p o c a l y p s e , as against one hundred and fifteen that are complete without it. Far more frequently than not the A p o c a l y p s e is found in Greek manuscripts b y itself a l o n e — s o tenuous was its connection with the Greek N e w Testament. Like the famous early cursive of the Bibliothèque Nationale, 3 familiarly known as 33, which Eichorn called " t h e queen of the cursives," Codex 2400 is typically Byzantine in its non-inclusion of the A p o c a l y p s e . T h e N e w Testament writings included in this codex are arranged in the normal Byzantine sequence, well known to students of the Greek Testament, but strange to those who know the literature in English translation only. The gospels stand Haseltine, Mr. G e o r g e E . D o w n i n g , and Dr. Ernest C. Colwell, I would make acknowledgment for much aid along the lines of their specialization. 2. T h e bibliography of Codex 2400, complete to date and arranged chronologically in order of publication, is as follows: Harold R . Wil lough by, The Rockefeller McCormick Manuscript of the Greek New Testament, in The University of Chicago Record, V o l . X I V , No. 3 (July, 1928), PP- 153-157. Harold R . W ilio ugh by, New Manuscript Acquisitions for Chicago, I The Rockefeller McCormick Testament, in The University of Chicago Magazine, V o l . X X I , No. 3 (January, 1929), pp. 128-133. Harold R . Willoughby, A Masterpiece of Byzantine Book Making, in Press Impressions, Vol. V I , No. 7 (February, 1929), pp. 1-6. Donald W . Riddle, The Rockefeller McCormick Manuscript, in The Journal of Biblical Literature, V o l . X L V I I I , Part 3 (1929), pp. 248-256. Harold R . W i l l o u g h b y , The Reconstruction of Lost Rockefeller McCormick Miniatures, in The Journal of Biblical Literature, V o l L I , Part 3 (1932), pp 179-188. Edgar j . G o o d s p e e d , The Rockefeller McCormick New Testament, I Facsimile, Chicago, T h e University of Chicago Press, 1932. Donald W . Riddle, The Rockefeller McCormick New Testament^ II The Text, Chicago, T h e University of Chicago Press 1932. Harold R . Willoughby, The Rockefeller McCormick New Testament, III The Miniatures, Chicago, T h e University of Chicago Press, 1932. 3. Paris, Bibliothèque Nationale, Grec 14.

i . My indebtedness to many friends and helpers, scholars both here and in E u r o p e , is so extensive that it cannot adequately be acknowledged in a single note. Nevertheless, it is a pleasure to express particular appreciation to those w h o have exhibited such enthusiastic interest in the interpretation of the miniatures that give C o d e x 2400 its great distinction. First of all, a c k n o w l e d g m e n t should be made of the generosity of the late Mrs. Edith Rockefeller McCormick, who afforded me free access to the c o d e x . T o Professor Edgar J. Goodspeed, the discoverer of the manuscript, 1 owe the opportunity of sharing in its publication. M. Gabriel Millet, Professor of Byzantine A r t in the C o l l è g e d e France, and his able assistant Mile, Sirarpie Der Nersessian have contributed valued j u d g ments on problematic points. T o M . Henri Omont, of the Bibliothèque Nationale, and the late Mr. Julius P. Gilson, of the British Museum, and Miss Belle D a Costa G r e e n e , of the Pierpont Morgan Library, I am warmly grateful for the opportunity of working with the great manuscripts in their charge. Miss Ethelwyn Manning, of the Frick A r t Reference Library, in New Y o r k City, has been very helpful in allowing the use of the extensive photograph collection in that institution. T o the Princeton Index of Christian A r t my debt is far greater than the interpretative and statistical liability reflected in these pages. T h e indebtedness to Professor Charles R . Morey and Professor A . M. Friend, Jr., is so great that it cannot be repaid. T h e most sincere appreciation is here recorded for their patient guidance in the field of Byzantine iconography. T o my colleagues at the University of Chicago, especially to Dr. Donald W . R i d d l e , Miss Elizabeth 3

THE ART

4

BULLETIN

first, in the usual Eastern arrangement, followed by Acts, then the collection of brief Catholic Epistles, and finally the extensive Pauline letters, including the Pastoral Epistles and ending with Philemon and Hebrews. The precedence thus given to the Catholic Epistles over the letters of Paul is characteristic of the Byzantines. Among the great codices this is the order of the fourth century Codex Vaticanus (B) 4 and the fifth century Codex Alexandrinus (A) 6 of the British Museum. In the fourth century Codex Sinaiticus (s),6 where the letters of Paul follow immediately after the gospels, the Catholic Epistles yet follow the Acts in sequence. Far more frequently than not in minuscule manuscripts the Pauline letters stand after the Catholic Epistles rather than before them. The table of contents for Codex 2400 is as follows : A blank fly leaf, f. 1. Illuminated section and canon tables, ff. 2-5. Frontispiece on purple parchment: Moses receiving the Law, f. 6v. The Gospels (Matthew, Mark, Luke, John), fif. 7-105. The Acts, ff. 106-37. The Catholic Epistles (James; I and II Peter; I, II, and III John ; Jude), if. 138-5 ir. The Epistles of Paul (Romans, I and II Corinthians, Galatians, Ephesians, Philippians, Colossians, I and II Thessalonians, I and II Timothy, Titus, Philemon, Hebrews), ff. 151-205. Miniatured frontispiece : King David as the Psalmist, f. 2o6v. Fly leaf with malediction, f. 207. The striking frontispiece at the end of the manuscript (Fig. 15), with its almost effaced portrait of King David, has an altogether exceptional importance in regard to content, because it suggests and all but proves that Codex 2400 originally included a psalter as well as the Byzantine Testament. Almost invariably the frontispiece to psalters of the aristocratic type was such a depiction of King David as a musician. It was the comparison of the defaced end piece in Codex 2400 with the psalter frontispiece in Suppl. Gr. 1 3 3 5 of the Bibliothèque Nationale (Fig. 16) that first suggested the engaging hypothesis of a lost psalter. This Bibliothèque manuscript, which was written at about the same time as the Chicago codex and in a closely related hand, further resembles the latter not only in the omission of the Apocalypse, but specifically in its section and chapter divisions as well. Thus Suppl. Gr. 1 3 3 5 , with its combination of New Testament and psalter, goes far toward demonstrating that Codex 2400 also once ended with the liturgical selections that constituted the Byzantine psalter. I.

THE

COVERS

The first impression made by Codex 2400 is one of unusual richness and splendor. For this impression the many miniatures, with their bright colors and gold backgrounds, and the two metal covers, with their designs done in deep repoussé, are chiefly responsible. 4.

Rome, Vatican Library, Vat. Gr. 1209.

5.

London, British Museum, Royal Ms. I.

6.

Leningrad, State Public Library, Ms. 259.

FIG. 3—Chicago, Potter Palmer Engraving

Collection:

of the Resurrection

FIG. 5—Paris, Bibl. Nat.:

FIG. 4—Grotta/errata, Library : Paleologan Book Cover

Coislin 200. Evangelist

Matthew

CODEX

2400 A N D I T S

MINIATURES

7

T h e c o v e r s are of b o a r d s , lined on the outside with purple velvet, o v e r which metal plates of

silvergilt have

been

affixed.

T e s t s made b y

Professor

William

D . H a r k i n s , G i b b s prizeman in C h e m i s t r y for the y e a r 1928, indicate that the silver b a s e is almost without alloy.

O n this the g o l d is laid as g o l d leaf.

O n the front c o v e r a central panel in high relief exhibits the Crucifixion ( F i g .

1),

s u r r o u n d e d b y a foliated frame on which the s y m b o l s of the e v a n g e l i s t s stand out in medallions.

T h e b a c k cover, which is without a b o r d e r , shows the R e s u r r e c t i o n in

e q u a l l y high relief ( F i g . 2). T h e s e two tableaux, the most crucial in the Christian passion d r a m a , the one e x h i b i t i n g the L o r d Christ as a d y i n g savior-god, and the other e x h i b i t i n g H i m as a rising savior-god, w e r e f r e q u e n t l y and

fittingly

employed

b y mediaeval b o o k m a k e r s as c o v e r designs for c o d i c e s of Christian scriptures. It is p e r f e c t l y o b v i o u s , h o w e v e r , that the two c o v e r plates do not b e l o n g in a single series.

T h e y differ entirely in a r r a n g e m e n t , i c o n o g r a p h y ,

style, and c h a r a c t e r of inscriptions.

together

composition,

F u n d a m e n t a l l y , the Crucifixion is a B y z a n t i n e

c o m p o s i t i o n , done from a model that w a s an i v o r y c a r v i n g — a t least so the technique of the rinceau b o r d e r w o u l d s u g g e s t .

It fits the C h i c a g o c o d e x so neatly that it may

have been e s p e c i a l l y e x e c u t e d to adorn this particular manuscript.

The Resurrection,

on the other hand, is a W e s t e r n and specifically Italian design, done from an e n g r a v i n g . It is v e r y d o u b t f u l if it was d e s i g n e d e s p e c i a l l y for this c o d e x . T h e absence of a border, the condition of the e d g e s , the ill-matched b o s s e s , and the mislocation of the outer pair of bosses, all indicate that the b a c k c o v e r has been a d a p t e d to its p r e s e n t use. T h e treatment of the Crucifixion on the front c o v e r is simple, s y m m e t r i c a l , and, on the whole, liturgical in character.

O n l y the V i r g i n and St. John are present on either

side ; not L o n g i n u s or S t e p h a t o n or other a c c e s s o r y

figures.

T h e Virgin is d e s i g n a t e d

the M o t h e r of G o d , M P 6T, and the B e l o v e d D i s c i p l e is named, 0 ArpOS] I£2[ANNH2] 0 8E0A0r[0S],

A b o v e the arms of the cross are m o u r n i n g angels.

T h e y are not

mere personifications of the sun and m o o n ; for the h e a v e n l y b o d i e s are not s u g g e s t e d b y either r a y s or crescent and, m o r e o v e r , the a n g e l s are definitely n a m e d b y initials, MI[XAHA] and r[AB]P[IHA]. A n a n i m a t e d Christ stands on the s u p p e d a n e u m of the cross, H i s feet apart and at an a n g l e .

T h e d r a p e d loin cloth h a n g s from H i s waist to H i s knees.

H i s shoulders

are well b e l o w the level of H i s hands and H i s b e a r d e d head is up, framed in a cruciform nimbus. O n the h e a d p i e c e of the cross the name Jesus Christ, IS XS, is inscribed. the cross on a scroll is the v e r y unusual inscription INBTA. usual titulus s u p p l i e d b y the g o s p e l

Above

O n e e x p e c t s here the

narratives, I(HSOrS) N(AZAPAIOS) B(ASIAETS)

T(QN) I(OrAAIQN). It is obvious, however, that the last letter of the C h i c a g o titulus is either alpha or delta.

T h e alpha a p p e a r i n g in the title of " St. John the D i v i n e " on this c o v e r plate

has a d i a g o n a l cross stroke rather than a horizontal stroke. a p p e a r in the inscription on the b a c k cover.

A l p h a s of the same sort

In fact, in c o n t e m p o r a r y

inscriptions

g e n e r a l l y alphas are done with d i a g o n a l cross strokes a n d deltas with c o r n e r pendants. The

combination is e x c e l l e n t l y

illustrated

on an i v o r y

M u s e u m r e c e n t l y published b y P r o f e s s o r G o l d s c h m i d t .

casket in

the

Cleveland

It w o u l d seem certain, there-

fore, that the last letter of the R o c k e f e l l e r M c C o r m i c k titulus is A.

THE

8

ART

BULLETIN

T h i s form of the titulus is not unique b y any means. On three well-known Byzantine monuments, two metal b o o k - c o v e r s in the M a r c i a n a and an ivory triptych in the B i b l i o t h è q u e Nationale, it appears spelled out in full ; ' so there can b e no doubt at all as to the m e a n i n g of A. It stands for A05H2 and it designates Christ as " K i n g of G l o r y . "

A gold

c r o s s reliquary formerly in the B u r n s collection also carries

precisely this inscription done in cloisonné

enamel.

T h i s reliquary is now owned b y

M r . A . S . D r e y of N e w Y o r k and Munich. P e r h a p s the b e s t known of the miniatures that record this inscription is the D e s c e n t from the Cross in Berlin Q

66

(fol.

256V).

F o r stark c o n t r a s t between inscription and illustration, however, the Matthean

Man

of Sorrows in L e n i n g r a d 1 0 5 — w h e r e the dead Christ is p o r t r a y e d nude against the arms of the cross, with 0 BA2IAEU2 TH2 A0EH2 plainly inscribed in large uncials above H i s h e a d — i s an unforgettable

picture.

Incidentally, in a cave below the cross, on the cover of Codex

2400,

is a skull recal-

ling the l e g e n d that Adam was buried beneath G o l g o t h a . 8 It is more than likely, however, that b a c k of this l e g e n d there is some c o n n e c t i o n between the skull and the p l a c e n a m e of G o l g o t h a itself ( G o l g o t h a = xpaviov, skull). T h e four apocalyptic b e a s t s in the medallions of the b o r d e r are definitely assigned to the four evangelists b y means of initials o b s c u r e l y i n c i s e d ; the distribution

being

that of J e r o m e , which was g e n e r a l l y a c c e p t e d in the W e s t . T h u s the eagle is assigned to IQfANNHS], the lion to MAP[K02], the angel to MATT[AI02], and the ox to A[0Y]K[A2]. W h i l e fundamental e l e m e n t s of the front c o v e r design are indubitably Byzantine, j u s t as c l e a r are the W e s t e r n influences there r e g i s t e r e d .

T h e definite attachment of

the apocalyptic b e a s t s to the evangelists as their symbols, the r e p l a c e m e n t of the sun and moon b y the mourning angels, the vivacity of the figure of Christ on the cross, and the posture of S t . J o h n are indubitable evidences of Latin influence.

Undoubtedly

this c o v e r was made in s o m e c e n t e r where the G r e e k E a s t and the Latin W e s t c a m e together. E x c e p t for the inscriptions, however, the design on the back c o v e r of C o d e x ( F i g . 2) is a W e s t e r n c o m p o s i t i o n .

2 4 0 0

T h e very theme of the resurrection was not a

popular one in the G r e e k E a s t , and the Byzantines applied the title Anastasis

(here

I ANA2TA2I2) not to the actual R e s u r r e c t i o n , but rather to the D e s c e n t to H a d e s , c o n c e i v e d as a s y m b o l i c equivalent. In the c e n t e r of the c o v e r design Christ steps out of a s a r c o p h a g u s , c a r r y i n g the b a n n e r of the R e s u r r e c t i o n

in H i s left hand, and raising H i s right hand so as to

display the g a p i n g nail wound. S i x a r m e d soldiers lie asleep in various postures a b o u t the sarcophagus, the heavy lid of which has been shoved off behind.

An undulating

l a n d s c a p e stretches to the rear with the three c r o s s e s of Calvary prominent on the right.

S u s p e n d e d in the sky above are obviously intrusive elements : a descending

angel at the left, a chapel at the right, and tufts of s h r u b b e r y toward the center. 7. The triptych in the Cabinet des Médailles of the Bibliothèque was published by Schlumberger in L'Epopèe Byzantine (Vol. 1, p. 17), and the two metal bookcovers in the Marciana are reproduced by Pasini in Il Tesoro di S. Marco (pis. X I I and XIII). For the Cleveland casket see A . Goldschmidt, Die Byzanlinischen Elfenbein Skulpturen I, pl. X L V I I .

For these decisive references and the solution of the titulus problem I am grateful to Mies M, Alison Frantz, of the Princeton Index of Christian Art. 8. S . C. Malan, The Book of Adam and Eve (London, 1S92), pp. 1 6 5 - 1 7 1 . See also Origen in his commentary on Matthew. Migne, Patr. Gr., Vol. 13, Col. 1777.

C O D E X 2400 A N D ITS M I N I A T U R E S

9

Identically this composition, line for line, with the exception of the inscriptions and the other intrusive elements above, is known as an anonymous Florentine engraving of the fifteenth century. It is no. 11 in a series of sixteen engravings illustrating the life of the Virgin and of Christ.9 A print is included in the private collection of Mr. Potter Palmer, of Chicago (Fig. 3), and other prints are to be found in various public collections of Europe. The series as a whole is done somewhat in the style of Alessio Baldovinetti, and the influence of Filippo Lippi is also noticeable. Indeed, at various times the series has been attributed to Lippi, and—without sufficient reason—to Botticelli and Nicoletto da Modena as well. The collocation of this fifteenth century Florentine engraving with the cover design on Codex 2400 provides the most definite evidence thus far isolated for the dating of the cover. A late fifteenth century date is at least possible, though early sixteenth century is doubtless more probable. A s to place, there is a tangible possibility that the small chapel, absent from the engraving but crowded into the upper right-hand corner of the cover, may provide a definite locus. It is an Italian chapel with a detached campanile. Gothic elements are obvious in the cusped arch over the door, the quatrefoil above, and the finial that tops the tower. The Arena chapel at Padua and the Portiuncula at Assisi are famous though somewhat dissimilar examples of such a small Italian chapel as this one. With a front cover fundamentally Byzantine yet perceptibly Western, and a back cover definitely Italian yet carrying a Greek inscription, the place of combination must be sought in some zone influenced by both Byzantium and Italy. The eastern littoral of the Adriatic was such an area. Among the eastern ports of the Adriatic none is more likely than Ragusa to have produced such treasures as these silvergilt covers. Also on the western shore of the Adriatic the city of Venice was carrying forward the Byzantine tradition in her own way during these centuries, and was producing very splendid metal work. It is certainly possible that one or both of the cover plates were designed there. II. DESCRIPTION

OF THE

MANUSCRIPT

In its external aspect Codex 2400 is a deep octavo volume measuring 6 cm. in thickness on the binding edge. At the present time it consists of 207 folios of thin parchment that average 15.5x20.5 cm. in size, the folios at the front of the codex being somewhat more contracted than those at the back. The actual height of the leaves at the outer edge ranges from 20.2 cm. to 20.8 cm. Originally the manuscript consisted of at least 221 folios (fly leaves included). No less than twenty of the original folios have disappeared at one time or another, leaving fourteen distinct gaps in the text. Of these lacunae five have been filled with six folios of foreign parchment—of very different quality from the original—on which portions of the excised text have been rewritten in hands notably less attractive than that of the original scribe. Nine gaps in the text remain unsupplied, representing a total of 9.

A . M. Hind, Catalog of Early Italian Engravings

in the British

Museum,

B. 1, pp. 102 ff.

IO

THE ART

BULLETIN

thirteen text folios that have disappeared. These, together with a missing frontispiece, account for the discrepancy between the 221 original folios and the 207 leaves that constitute the codex today. The manuscript was written on a superior quality of parchment, distinguished for its softness and thinness. Because of these very excellencies, however, it has not stood the rough usage of time very successfully. In the front of the codex the leaves are shrunken and brittle, and considerably worn on the lower, outer edges, where they have been subjected to much thumbing. The later folios, however, are generally more alive and flexible. It is perfectly patent from the condition of the leaves that the gospel corpus was the most used section of this N e w Testament; Acts, too, attracted considerable attention, doubtless because of its miniatures; but the epistles were comparatively ignored. The codex throughout has suffered much from water, which has caused the parchment to contract and wrinkle, and has even smudged the text in spots. A plausible story told b y a former owner may account for this condition. A t one time in its history, he said, the manuscript acquired the reputation of being a holy book, charged with magical properties for the cure of diseases. The curative process was the hydropathic treatment of both the codex and the patient. W a t e r was poured on the c o d e x ; then the patient drank the water. Whatever were the consequences for the patient, the manuscript suffered badly as a result of this maltreatment. U n d e r the circumstances, it is surprising that almost none of the text was rendered illegible by the wetting. B y almost a miracle the miniatures escaped injury altogether. T h e y were untouched by the water; and, except for the rubbing and flaking that are normal for Byzantine miniatures, they are not in notably bad condition at the present time. A particularly attractive feature of Codex 2400 is the well adjusted proportion between the margins and the panel of writing which they frame. The average measurements of this panel are 1 5 x 1 0 cm. The outer and lower margins approximate each other in width, and the upper and inner margins tend 10 do the same. There is no indication that the manuscript has been trimmed. T h e system of marginal ruling is also interesting. Vertical double lines were used to bound the text on either side. In addition, another pair of lines, also vertical, was drawn through the middle of the wide outer margin. Y e t other double lines divided unequally the upper and lower margins as well. A special refinement of the manuscript is the decorative handling of the text to form patterned panels. On fol. 66r a cross is thus formed b y a half page of text (Fig. 6). A t the end of the Gospel of Matthew the text tapers off to a point formed by the single letters of the " A m e n " at the end of the gospel. Similar tapering, though less pointed, is pronounced on other pages, usually just before a miniature at the top of the succeeding page. 10 The Lucan genealogy in this manuscript is attractively ordered in four narrow columns (ff. 6 i v a n d 62r. See Fig. 7). One cannot but wonder, of course, what may have been the arrangement of the Matthean genealogy as the original scribe wrote it.

10.

O b s e r v e particularly the text a r r a n g e m e n t on

the f o l l o w i n g f o l i o s : 36V, 54r, o 6 r , i37r.

55r, 68r, 6gr, gor,

FIG. 8—Chicago, RockefellerMcCoi-mick Benediction

Collection: Byzantine

of the Apostles ; Evangelist

FIG. 9 — P a r i s , Bibl. Nat.:

Coislin 200.

New

John

Evangelist

John

Testament.

CODEX

2400 A N D I T S

MINIATURES

13

Two closely related manuscripts in the B i b l i o t h è q u e Nationale exhibit a similar scribal fondness for decorative designs rendered in text. Coislin 200, which was penned b y the same scribe as the Chicago codex, has such adornments at the ends of various b o o k s . " In Suppl. Gr. 1 3 3 5 — w r i t t e n at about the same time and in a hand not d i s s i m i l a r — t h e textual designs are more e l a b o r a t e ; for this fastidious scribe did not hesitate to affect even the patriarchal cross on occasion. 1 2 T h e text of Codex 2400 is written in single columns that measure, as indicated, 1 0 x 1 5 cm. In the gospels section the text runs normally to 42 lines p e r page, while in the A c t s - E p i s t l e s section it r e a c h e s only 36 lines per page. T h e ink is black and g l o s s y — o f exceptional quality, like the p a r c h m e n t on which it is w r i t t e n — s o that, as noted above, even the pouring of water on the manuscript has not rendered the writing illegible. O n e of the most attractive features of the c o d e x is the style of its script. It is written in a very clear, very fine, very beautiful cursive hand, which for all its minuteness is very e a s y to read. W h i l e characterized b y g r e a t c o h e r e n c e and regularity it has also great individuality and distinction. D i a g o n a l strokes are accentuated, c r o s s curves are flattened, and loops are executed with a flourish ( F i g . 6). 1 3 Identically this hand is seen in such well-known codices as Coislin 200 of the B i b l i o t h è q u e Nationale and L a u r a B 26 on Mt. Athos. F o u r t e e n various hands altogether have been recognized in the Chicago manuscript as a whole. 1 4 In the main, these filled in different lacunae in the text, added lection notes and section headings, a few marginal glosses and fewer corrections, notes at the end of the second and fourth gospels and R o m a n s , a date at the beginning of the manuscript and a malediction at the end. A few c o r r e c t i o n s by the original scribe have been n o t e d ; but it is conspicuous that no diortholes, or c o r r e c t o r , improved his work. Of these various later hands P r o f e s s o r E d g a r J . G o o d s p e e d has identified the hand that filled in the earlier g a p s in the text (ff. 7, 8, and 60) as a script of the fourteenth century, about a hundred years only after the manuscript itself was written. P r o f e s s o r G o o d s p e e d is also confident that another scribe, who added a note at the end of Mark, supplied a folio in L u k e , filled in the lacunae in the epistles, wrote an 5jto6eais to J a m e s , and chapter titles to H e b r e w s (ff. 5 5 r , 7 0 , 1 3 7 V , 1 4 6 , 151, 2 0 5 V , 2o6r), did his work in the fifteenth century. 1 3 Titles, whether of chapters or of books, are done in gold over red, the chapter titles b e i n g disposed in one of the outer margins. Gospel titles are rendered impressively in a combination of uncial and cursive letters, the uncials predominating, and the usual form of gospel title b e i n g used ( F i g . 8). T h i s ornamental text, t o g e t h e r with the portrait headpiece above, invariably constitutes a decorative panel of g r e a t 11.

S e e ff. 42V, 691", 109V, and 300V of Coislin 200.

12.

In Suppl. G r . 1335 see the following p a g e s :

4 2 v , 651-, 6 5 V , 1 0 3 V , 1 2 9 V , 1 6 7 V , 1 7 3 1 - , i 8 4 r , 1 8 5 V , 2941-,

327V, 3 3 3 r , 334V. 13. F o r a detailed study of the paleography of the manuscript see The Rockefeller McCormick New Testament, V o l . I I , The Text, by Donald W . Riddle (Chicago, 1932).

14. Donald W . Riddle in The Journal of Biblical Literature, V o l . X L V I I I , Part 3 (1929), pp. 250 f. 15. F o r the identification o f the different later hands see Professor G o o d s p e e d ' s Introduction in The Rockefeller McCormick New Testament, V o l . I , Facsimile (Chicago, 1932).

14

THE ART

BULLETIN

charm and consistency. At the conclusion of the gospel the xiXo; is written in a flourished minuscule hand (Fig. 8). In the epistle titles the cursive letters predominate over the semi-uncials. Ammonian section numbers are registered, also in gold over red, in the margins; but the Eusebian canon numbers are conspicuously absent. This is the more remarkable since the full series of Eusebian canons are brilliantly illuminated at the beginning of the codex. Errors in the registering of the section numbers are also prominent. The last series in Luke ends with 242, when it should be 342. Similarly the final series in John concludes with 1 3 2 , instead of 232. Here, again, a diorthotes should have corrected the original copy. In general, both the chapter and section divisions in this manuscript parallel the divisions in related manuscripts, Laura B 26 and Suppl. Gr, 1 3 3 5 in Paris. In the praxapostolos section of the Chicago Testament vermilion liturgical notes enliven the margins, as do the gold titles in the gospels. The leaves of the codex are bound together today in twenty-nine gatherings of five to eight folios each, with a concluding union of folios that are, however, patched together. When the codex was made the standard gatherings were ternions and quaternions, the former appearing toward the end of a book when the amount of space required by the text so dictated. Also at the end of the fourth gospel and again at the beginning of Acts, there were quires complete with only seven folios, the odd leaf in each instance being fixed in with a stub. At the end of Luke, apparently, an extra union was added, of which only one folio remains. These peculiarities would seem to indicate a scarcity of the finest quality of parchment at the time the manuscript was being written. S o far as the quire numbers remain, they may be read in the upper right corner of the first recto of the gathering. The sequence of quires, with a notation of both present and original contents, is as follows: Preliminary leaves, six in number (ff. 1-6), consisting of a blank fly leaf, a binion on which the ten canon tables are illuminated (complete), and an inserted frontispiece of purple parchment that was originally the first leaf of a union of purple folios. Quire 1, of nine leaves (ff. 7-15), originally also an enlarged quaternion, of which the first leaf, containing the portrait of the evangelist, and the second leaf, carrying the gospel title and Matt, i: 1-25, are now missing. This lacuna was supplied as early as the fourteenth century by an inserted union of folios (ff. 7 and 8), on which the lost text was rewritten, the first eight verses being rendered in duplicate on the second recto. Quire 2, of seven leaves (ff. 16-22), numbered B, also a quaternion originally, of which the second leaf, containing Matt. ix:2o-38, is now missing. Quire 3, also of seven leaves (ff. 23-29), numbered r, originally a quaternion, from which the fifth leaf has disappeared and with it Matt. xx:2ob-xxi: 15 a. Quire 4, complete with six leaves (ff. 30-35), numbered A. The Gospel of Matthew concludes with this gathering. Quire 5, complete with eight leaves (ff. 36-43), numbered e. Quire 6, complete with six leaves (ff. 44-49), numbered s.

CODEX

2400 A N D I T S

MINIATURES

Q u i r e 7, c o m p l e t e with s i x l e a v e s (ff. 50-55), n u m b e r e d

15 H e r e the G o s p e l

of

Mark ends. Quire 8, of e i g h t l e a v e s (ff. 56-63), n u m b e r e d i).

T h e fifth leaf of this g a t h e r i n g

w a s inserted in the f o u r t e e n t h c e n t u r y (with the text of L u k e ii:27b-iii :9a), to take the p l a c e of a lost o r i g i n a l folio. Quire 9, of s e v e n l e a v e s (ff. 64-70), r e p r e s e n t i n g an o r i g i n a l quaternion f r o m which the first and last f o l i o s w e r e r e m o v e d at an e a r l y date. the text of L u k e i v : 4 2 b - v . 3 3 a , is still missing.

T h e first leaf, which c a r r i e d

T h e last leaf only (f. 70) with the

text of L u k e i x : 3 6 b - x : 6 a w a s later s u p p l i e d in a hand p r e s u m a b l y of the fifteenth century. Q u i r e 1 0 , c o m p l e t e with s i x l e a v e s (fif. 7 1 - 7 6 ) , n u m b e r e d 0. Q u i r e 1 1 , of eight l e a v e s (ff. 77-84), o n c e n u m b e r e d t, but later c h a n g e d to ia. T h e o r i g i n a l eighth folio of this g a t h e r i n g is m i s s i n g and with it the text of L u k e x x i i i : 3 9 xxiv:2ia.

A t the e n d of the quire a s constituted at p r e s e n t an e x t r a leaf (f. 84) is

attached.

A t the outset this w a s the s e c o n d folio of an e x t r a union that w a s n e c e s s a r y

in o r d e r to c o n c l u d e the text of L u k e . Q u i r e 1 2 , of s i x l e a v e s (ff. 85-90), o r i g i n a l l y a quaternion of which the f o u r t h and fifth folios (John ii : 3b-iv : 1 0 a ) are now l a c k i n g . Quire 1 3 , c o m p l e t e with eight l e a v e s (ff. 9 1 - 9 8 ) , n u m b e r e d iji but altered to 18. Quire 1 4 , c o m p l e t e with o n l y s e v e n l e a v e s (ff. 9 9 - 1 0 5 ) , n u m b e r e d c h a n g e d to te.

which has b e e n

F o l . 4 at the c e n t e r of this d e f e c t i v e quaternion h a s a stub.

With

this g a t h e r i n g the text of the g o s p e l s is b r o u g h t to a c o n c l u s i o n . Q u i r e 1 5 , of s i x l e a v e s (ff. 1 0 6 - 1 1 1 ) , l a c k i n g the first leaf of the o r i g i n a l g a t h e r i n g ( A c t s i : i - i o a ) , i n c l u d i n g the m i n i a t u r e d h e a d p i e c e of A c t s . F o l . 2 in this g a t h e r i n g is a t t a c h e d with a stub, s h o w i n g that this q u i r e a l s o w a s o r i g i n a l l y a d e f e c t i v e quaternion c o m p l e t e with o n l y s e v e n l e a v e s . Quire 1 6 , of s i x l e a v e s (ff. 1 1 2 - 1 1 7 ) .

A t the outset, h o w e v e r , this w a s a quaternion,

of which the fourth and fifth l e a v e s are now l a c k i n g ( A c t s v i i : 58b-ix : 3). Quire

1 7 , c o m p l e t e with six l e a v e s (ff. 1 1 8 - 1 2 3 ) , n u m b e r e d r , thus

emphasizing

the f a c t that a new quire n u m b e r i n g b e g i n s with A c t s . Q u i r e 1 8 , of s i x l e a v e s (ff. 1 2 4 - 1 2 9 ) , n u m b e r e d A.

This gathering was a quaternion;

but fol. 3 ( A c t s x v i : 39b-xvii : 22) and fol. 6 (Acts xix : 2 5 b - x x : 7a) h a v e d i s a p p e a r e d . Quire 1 9 , of e i g h t l e a v e s (ff. 1 3 0 - 1 3 7 ) , c o m p l e t e . T h e B o o k of A c t s is c o n c l u d e d b y this g a t h e r i n g . Quire 20, of e i g h t l e a v e s (ff. 1 3 8 - 1 4 5 ) , c o m p l e t e , n u m b e r e d ?. Quire 2 1 , of five l e a v e s (ff. 1 4 6 - 1 5 0 ) .

T h e first leaf (f. 1 4 6 ) w a s inserted a p p a r e n t l y

in the fifteenth c e n t u r y to s u p p l y the text II P e t e r iii : n b - i 8 a n d I J o h n i : i - 9 a . P r e s u m a b l y this g a t h e r i n g w a s o r i g i n a l l y a c o m p l e t e ternion.

If s o it c a r r i e d on its

last folio, now lost, the conclusion of J u d e ( 1 4 ^ 2 5 ) , at p r e s e n t s u p p l i e d in a fifteenth c e n t u r y h a n d on fol. 1 5 1 , at the b e g i n n i n g of the next g a t h e r i n g . Quire 2 2 , of s i x l e a v e s (ff. 1 5 1 - 1 5 6 ) , the first leaf b e i n g i n s e r t e d in the

fifteenth

c e n t u r y to take the p l a c e of an original folio which c a r r i e d a p o r t r a i t of P a u l a n d the o p e n i n g v e r s e s of R o m a n s (1 : i-26a). Q u i r e 2 3 , c o m p l e t e with eight l e a v e s (ff. 1 5 7 - 1 6 4 ) . Q u i r e 24, c o m p l e t e with eight l e a v e s (ff. 1 6 5 - 1 7 2 ) .

i6

THE ART

BULLETIN

Quire 25, complete with eight leaves (ff. 173-180). Quire 26, complete with eight leaves (ff. 181-188). Quire 27, complete with eight leaves (ff. 189-196). Quire 28, complete with eight leaves (ff. 197-204). Quire 29, of two leaves (ff. 205 and 206) patched together, the first concluding Hebrews, and the second carrying a defaced miniature on its verso. Final fly leaf, very wrinkled, with a malediction difficult to decipher inscribed upon it. III.

THE

TEXT

The study of the structure and paleography of Codex 2400 has shown that fourteen gaps have been made in the text since it was originally written, representing a total of eighteen text folios. Within a century after the manuscript was penned two of these folios had been abstracted and the gaps thus made had been supplied in parchment and script of the fourteenth century, three folios being inserted at this time to make good the missing text (Matt, i : 1-25 ; Luke ii : 27b-iii : 9a). Within another century three more folios had been supplied, on which text was transcribed in a fifteenth century hand, to fill in three more gaps in the text (Luke ix : 36b-x : 6a; II Peter iii : n b - 1 8 ; I John i : i-ga; Jude 14-25; Rom. i : i-26a). In all of these cases differences in parchment, style of writing, and quality of text make the later insertions perfectly obvious. Nine lacunae in the text, however, remain unfilled at the present time. These are in the gospels—with the exception of Mark—and in Acts also. The missing passages, which represent thirteen lost folios, are as follows: Matt, ix : 20-38; Matt, xx : 20bx x i : i 5 a ; Luke iv : 42b-v : 3 3 a ; Luke xxiii : 39 xxiv 2 i a ; John i i : 3b-iv : 1 0 a ; Acts i : i - i o a ; Acts vii : 58b-ix : 3 ; Acts xvi : 39b-xvii : 22 ; Acts x i x : 25b-xx : 7a. Patently Acts has suffered more than any other single book from excisions. An altogether singular hiatus in the text is at John viii : 21-59 (fol. 93V, fifth line), including Jesus' argument with the Jews in the temple about descent from Abraham. Occurring as it does within a single line, there is nothing to indicate that the scribe was conscious of any omission at this point. Under the circumstances it seems most probable that his exemplar was here defective to the extent of a whole folio. Again the absence of a diorthotes to correct the error is attested. There are in the New Testament a few outstanding passages of crucial textual significance, to which the student always turns first of all, in the hope of finding that a given manuscript will prove to be of primary importance. Chief among them are the conclusion of Mark's gospel, and the pericope about the woman caught in adultery (John vii : 53-viii : 11). A t the end of the second gospel Codex 2400 has the socalled long conclusion (Mark xvi : 9-20, fol. 55r), in the same form as does Codex Alexandrinus (A) and Efraem (C) and Bezae (D) and, indeed, all late uncials and all cursives. This was to be expected. It was not, however, to be expected that the beginning of the pericope in John would be marked by a cross in such a way as to indicate a peculiarity as to its status (John vii : 53, fol. 93r). What is very surprising, moreover, is the ending of Romans in Codex 2400. The long doxology of Rom. xvi : 25-27 is in this manuscript located at the end of chapter

C O D E X 2400 A N D I T S

MINIATURES

17

14 (fol. i6or) as it is in Codex Angelicus (L2) of the ninth century and in over 200 Greek cursives; and also as it was originally in the Latin Codex Amiatinus (A) of the eighth century—so the summary of its contents proves. Chapter 16 in the Chicago codex ends instead with the brief benediction repeated from Rom. xvi : 20, " T h e blessing of our Lord Jesus be with you all, A m e n " (fol. i6ir). The manuscript thus adds its modicum to the engaging hypothesis that there once existed a short recension of Paul's Roman letter, from which chapters 15 and 16 were lacking. Textually Codex 2400 has proved to be far more varied and interesting and significant than was anticipated at the outset. Dr. Riddle's results, which are published as Volume II of the Rockefeller McCormick publications, demonstrate that in numbers of instances the text departs from the Syrian type that is usual in minuscule manuscripts. In the gospels of Mark and Luke and John and more particularly in the Pauline Epistles a considerable proportion of its readings agree with the better witnesses found among the older uncials. Agreements are numerous with the important Koridethi Codex (0), and with the readings published by Professor Lake as typical of " Family 6 . " Perhaps it is not too much to say that Codex 2400 presents a weakened form of the Caesarean text that is occupying scholarly attention prominently at present. In the gospels section the textual affinities of the manuscript are with Trinity College B . 1 0 . 1 6 , Burney 20 and 21 of the British Museum, and the Pickering manuscript, the last two being the work of the famous scribe, Theodore Hagiopetrites. In Acts the text of Codex 2400 is without distinction except between iii:2oand v : 29, where the text relates to a considerable family of manuscripts. 16 The Pauline Epistles present the best text; and here the correlation is with the famous Leicester codex and Leningrad 1 0 1 . It is a point of special interest since it involves a familiar, liturgical expression, that this codex is one of a very few manuscripts to omit the words " Take, eat! " from the Pauline paradosis of the Lord's Supper (I Cor. xi: 24, fol. i67r).

IV.

DATE

AND

PROVENANCE

The dating of Codex 2400 hinges on a comparison of it with Coislin 200 of the Bibliothèque Nationale, a thirteenth century New Testament (minus the Apocalypse) miniatured with author portraits only. Incidentally Coislin 200 is a very famous manuscript, well known to paleographers and students of New Testament text. Apparently it was the manuscript designated 0 in Stephanus' influential folio edition of 1 550. A comparison of the two manuscripts throughout—in photographs, photostats, and in the originals—has established conclusively that both were penned by the same scribe. The distinctive features of the two manuscripts from the point of view of paleography are the following (compare Fig. 8 with Fig. 9): 1. The general coherence of the script and the clear definition of the line of writing in contrast to the intervening space.

16.

T h e G r e g o r y numbers for this family of rela-

ted manuscripts are: 479> 483, and 642.

1 0 1 , 102, 201,

206, 3 1 9 , 462,

THE ART

8 2.

BULLETIN

The beauty, g r a c e , neatness, sureness, and regularity of the script.

3. T h e detailed resemblance of individual letters; notably the united double lambdas with their sharp diagonal strokes, the flourish on the cursive delta, the psi with the flatness of its cross curve, the broad kappas, and the upper loop of the phi. 4. T h e balanced proportion between uncial, semi-uncial, and cursive letters in book headings and the initial lines of gospels. Since both manuscripts exhibit alike these paleographic peculiarities it is certain that the two were written b y the same scribe. Furthermore, distinctive decorative motifs are identical in the two codices. This is shown most strikingly by a comparison of author portraits in the Chicago codex with gospel headings in Coislin 200. (Again compare F i g . 8 with F i g . 9.) In the latter manuscript the gospels are p r e f a c e d by full page evangelist portraits and also b y square, decorative headpieces above the text. In Codex 2400 the evangelist portraits are set within the square headpieces as in frames. Detailed comparison in the case of each evangelist proves the equation, that the Coislin portrait, plus the Coislin border as a frame, equals the framed portrait of the Rockefeller McCormick evangelist. T o be specific, the analysis of the particular attention to the following portrait frame in the Chicago c o d e x : 1. T h e nearly square shape of the 2. T h e foliated decorations within trefoils outside.

fourth gospel headpiece in Coislin 200 calls elements that conspicuously characterize the design. the design; cinquefoils set in medallions and

3. Simple corner finials, slanting above and upright below. Iconographically, also, the two series of author portraits are v e r y similar throughout, and are characterized b y features typical of Byzantine art in its late development. In each case the fourth evangelist is a venerable, stoop-shouldered figure, with a long beard, leaning heavily over his half-opened codex. ( A g a i n compare F i g . 8 with F i g . 9.) L i k e w i s e in the epistle cycles the supposed apostolic authors are shown full front as f r a m e d half figures, each with a book or roll in his left hand, and with his right hand raised in benediction. F o r these combined likenesses—iconographic, stylistic, and p a l e o g r a p h i c — t h e reasonable explanation would seem to be that the two codices were miniatured in the same scriptorium as well as written b y the same scribe. It is definitely known where and when this writing was done. In 1 2 6 9 Michael V I I I Paleologus, E m p e r o r of Byzantium, sent Coislin 200 as a gift book to L o u i s I X , K i n g of F r a n c e — t h e St. L o u i s who built Sainte-Chapelle next the royal library to enshrine the relics of the Crucifixion, and who sailed on a tragic crusade from A i g u e s Mortes in the year 1 2 7 0 to his own death in Carthage. On the frontispiece to the first gospel in Coislin 200 there is a sprawling colophon (fol. 3r, F i g . 5), generally believed to be the presentation autograph of Michael Paleologus himself, in which the emperor is n a m e d : " M i c h a e l in Christ G o d , the faithful K i n g and E m p e r o r of the R o m a n s ; D u c a s , A n g e l u s , Comnenus, P a l a e o l o g u s . " On the recto of the same folio are Latin colophons, written in a thirteenth century hand, which record the negotiations of 1 2 6 9 , looking toward the unification of Christendom, that caused the sending of the imperial gift book to S t . L o u i s . Paleographers without exception, from Montfaucon to Gre-

C O D E X 2400 A N D I T S

MINIATURES

19

gory and von Soden, have assigned the writing of Coislin 200 to the thirteenth century, and most of them state explicitly that it was written specifically on the order of Michael Paleologus himself ; " X I I I , ut videtur, saeculo jussu Michaelis Palaeologi descriptus," as Montfaucon summarizes in his Bibliotkeca CoislinianaP This dating of Coislin 200 is strikingly confirmed by a comparison of Grec 1 1 7 of the Bibliothèque Nationale, written in a closely related script and dated A. D. 1262. On the other hand, the well-known Laura B 26 on Mt. Athos, which was apparently written by the same scribe as Coislin 200, carries a colophon (fol. 268r) dating it in the year 1084 and in the reign of Alexius I Comnenus, nearly two centuries before the period of Michael Paleologus: "Written in the reign of the most pious and most Christian Lord Alexius the Great, Ducas Augustus Comnenus, in the year 6592, solar cycle 9, lunar cycle 18, 8th indiction, the first day of Lent January 3 1 , legal Easter March 25, Christian Easter March 28, the fast of the Holy Apostles day 36." 1 8 Assistant Professor E . C. Colwell in looking up the indiction discovered that not one of the data cited was correct for the year 1084 except the lunar cycle. Moreover, the only year for which all the given items would be correct was the impossible year 1445. Accordingly, the date colophon of Laura B 26 may be considered to negative its own evidence by its very patent errors. It seems probable that the colophon, as it now appears, was recopied, perhaps incorrectly, in the thirteenth century from an eleventh century exemplar. While a scribe could scarcely make so many mistakes about the year in which he was actually writing, a later copyist might easily make such blunders about an earlier date. The miniatures in Laura B 26 confirm this hypothesis. For them the year 1084 cannot possibly be the correct date. The thirteenth century would be much more probable. On the whole, they are not far in time from the Rockefeller McCormick miniatures. It seems clear, then, that Laura B 26 must be regarded as a thirteenth century product, copied from an eleventh century original. 19 The consensus of evidence, iconographie, stylistic, and paléographie, from these related manuscripts points to the early Paleologan period as the time of production, and to the imperial scriptorium of Constantinople as the place. There is the very highest probability that Codex 2400, like Coislin 200, was produced in the palatine scriptorium of Michael Paleologus and at his order. In the total history of the Byzantine empire the palace-scriptorium of the emperors 17. Bern. Montfaucon, Bibliotkeca Coisliniana {Paris, 1715), p. 250. Compare also the following : J . J . Wetstein, Novum Testamentum Graecum (Amsterdam, 1751), Vol. I, p. 49. J . B. Silvestre, Universal Paleography (Translated by F. Madden, London, 1849), Vol. I, p. 122. F. H. A. Scrivener, A Plain Introduction to the Criticism of the New Testament (London, 1861), p. 146. Berger de Xivrey, Notice d'un Ms grec du XIIIe siècle (Paris, 1863). H. Bordier, Description des Peintures.... dans les manuscrits grecs de la Bibliothèque Nationale (Paris, 1883), p. 226.

C. R. Gregory, Prolegomena to Tischendorf's Novum Testamentum Graece, Ed. VIII (Leipzig, 1890), Vol. 3, p. 471. C. R. Gregory, Canon and Text of the New Testament (New York, 1907), p. 373. H. von Soden, Die Schriften des Neuen Testaments (Berlin, 1902-10), Vol. i, Part 1 , p. iro (8355). 18. For a transcript of this colophon the author is indebted to Archbishop Sophronios Eustratiades, editor (with the monks Spyridon and Arcadios) of the invaluable catalogues of Athonite manuscripts at the Laura and Vatopedi. 19. This explanation of the date colophon in Laura B 26 was first suggested to me by Professor A . M. Friend, Jr., of Princeton.

20

THE

w a s an o u t s t a n d i n g institution.

ART Under

BULLETIN Constantine VII Porphyrogenitus

it

was

l o c a t e d in the S a c r e d P a l a c e . H e r e the u s u r p e r R o m a n u s I L e c a p e n u s c o n f i n e d the e m p e r o r himself f r o m g i g to g 4 4 , a n d here the e m p e r o r b u s i e d himself m o s t h a p p i l y with his w r i t i n g s on c o u r t c e r e m o n i a l .

C o n s t a n t i n e w a s b o t h a c a l l i g r a p h e r and a

miniaturist, a n d he e v e n " c o r r e c t e d " miniatures b y m o d i f y i n g their i c o n o g r a p h y . U n d e r B a s i l II B u l g a r o c t o n u s (976-1025) the i m p e r i a l s c r i p t o r i u m w a s m o v e d to the P a l a c e of B l a c h e r n a e .

H e r e the most f a m o u s of B y z a n t i n e m e n o l o g i a , the

c o d e x n o w in the Vatican, 2 0 w a s m i n i a t u r e d for B a s i l II. e m p l o y e d on the m a s t e r p i e c e .

great were

E i g h t of them s i g n e d their w o r k — a s B y z a n t i n e minia-

turists d i d not u s u a l l y t r o u b l e to do. t h e m s e l v e s as artists " o f

A n u m b e r of artists

T w o of t h e m , M i c h a e l a n d S i m e o n , d e s c r i b e d

Blachernae."

W h e n , in 1 2 6 1 , M i c h a e l P a l e o l o g u s r e c a p t u r e d C o n s t a n t i n o p l e f r o m the L a t i n s the palatine s c r i p t o r i u m w a s r e e s t a b l i s h e d at B l a c h e r n a e , a n d d o w n to the t r a g i c e n d of the P a l e o l o g a n p e r i o d it c o n t i n u e d to p r o d u c e m a s t e r p i e c e s of the b o o k m a k e r ' s art. A m o n g the earliest of t h e s e m a s t e r p r o d u c t s w e r e the St. L o u i s g i f t b o o k a n d the Chicago N e w Testament.

It is a s h r e w d g u e s s that if M i c h a e l

Paleologus

thought

it w o r t h w h i l e to s e n d to L o u i s I X of F r a n c e s u c h a c o m p a r a t i v e l y c r u d e p i e c e of w o r k as Coislin 200, he w o u l d b e a l t o g e t h e r l i k e l y to k e e p for himself s u c h a s p l e n d i d p i e c e of w o r k as C o d e x

2400.

T h i s s u g g e s t i o n is e x a c t l y in line with the k n o w n

b i b l i o p h i l e h a b i t s of this f a s c i n a t i n g e m p e r o r , a s a t t e s t e d in a n u m b e r of m a n u s c r i p t colophons. T h e B i b l i o t h è q u e N a t i o n a l e t o d a y p o s s e s s e s y e t a n o t h e r c o d e x , G r e c 1 1 1 5 , written on the o r d e r of M i c h a e l P a l e o l o g u s e x p r e s s l y for the i m p e r i a l l i b r a r y . It is a c o l l e c t i o n of a n o n y m o u s t h e o l o g i c a l tracts, t r a n s c r i b e d in the y e a r 1 2 7 6 b y the s c r i b e L e o Cinnamus.

T h e c o l o p h o n on fol. 3o6r r e a d s in p a r t : " T h i s b o o k w a s written b y the h a n d

of L e o C i n n a m u s , and c o m p l e t e d with G o d ' s help on the f o u r t e e n t h d a y of M a r c h , the fourth i n d i c t i o n , the s e v e n t h d a y in the y e a r 6784, at the i m p e r i a l c o m m a n d of the most p i o u s and faithful and G o d - c r o w n e d rulers, the L o r d M i c h a e l D u c a s

Angelus

C o m n e n u s P a l e o l o g u s , the n e w C o n s t a n t i n e , and the most p i o u s A u g u s t a T h e o d o r a . . . . and it w a s d e p o s i t e d in the i m p e r i a l l i b r a r y . " 2 1 O n e of the m o s t i m p r e s s i v e r e m n a n t s of the i m p e r i a l l i b r a r y extant

t o d a y is a

c o l l e c t i o n of t r e a t i s e s b y the E m p e r o r M a n u e l II P a l e o l o g u s ( i 3 g i - i 4 2 5 ) p r e s e r v e d in the a b b e y l i b r a r y at G r o t t a f e r r a t a , 2 2 w h e r e the G r e e k l i t u r g y is still used, and w h e r e a m o n a s t i c s c r i p t o r i u m is still m a i n t a i n e d .

T h i s i m p e r i a l c o d e x is

covered

with b l u e silk on w h i c h c h a r a c t e r i s t i c P a l e o l o g a n d e s i g n s are w o r k e d with t h r e a d s of silver ( F i g . 4).28 In the c e n t e r is the d o u b l e h e a d e d P a l e o l o g a n e a g l e and in the c o r n e r m e d a l l i o n s the P a l e o l o g u s m o n o g r a m stands o u t .

It is t o d a y a u n i q u e

Byzantine

b o o k c o v e r , and it g i v e s a s t r i k i n g i m p r e s s i o n of the s u m p t u o u s v o l u m e s that w e n t into the m a k i n g of the l i b r a r y of the P a l e o l o g i .

A m o n g s u c h s p l e n d i d c o d i c e s , the

p r o d u c t s of a palatine s c r i p t o r i u m and the t r e a s u r e s of an i m p e r i a l l i b r a r y , C o d e x 2400 o r i g i n a l l y held a h i g h p l a c e . 20.

Rome, Vatican Gr. 1613.

21. H . Omont, Facsimiles des Manuscrits Grecs datés de la Bibliothèque Nationale (Paris, 1929), p. 13, pl. L X I I .

22. 2J.

Grottaferrata, Z. Ç. 1.

. Mufioz, L'Art Byzantin a l'Exposition: Grottaferrata, fig. 101, pp. 144 f. A

de

C O D E X 2400 A N D I T S

V. THE

MINIATURES

23

MINIATURES

The great distinction of Codex 2400 relates to the number and character of its miniatures. These are found in abundance not only throughout the gospels as text illustrations but also in Acts, with scattered author portraits in the epistles section. There are in the manuscript today an even ninety miniatures, with eight richly decorated canon tables at the beginning of the codex. Altogether, then, there are nearly one hundred brilliantly illuminated pages in Codex 2400. As is usual in East Christian manuscripts, the miniatures (i. e., the pictures) and the illuminations (i.e., the decorative designs) are carefully differentiated from each other, the miniatures being framed very simply in single lines or plain borders, and the illuminated canon tables combining architectural motifs with plant and bird forms only. Three evangelist portraits, however, have wide frames, heavily gilded and even intricate in patterned ornamentation (color plate and Figs. 8, 14); and three apostle portraits, also, have less wide, but very colorful borders (Figs. 55-57). Of the ninety miniatures, six are author portraits, two are frontispieces, and the remaining eighty-two are scenes illustrative of particular texts. The miniatures are distributed as follows: in Matthew twenty-four, in Mark seventeen, in Luke nineteen, in John twelve, in Acts thirteen, in the Catholic Epistles three. B y a considerable margin the first gospel has the most miniatures. Of the books that are illustrated in this New Testament, Mark is the only one that has lost none of its original miniatures. Matthew, Luke, and Acts have all lost heavily in illustrations—in round numbers, at least half a dozen miniatures apiece. Losses from the fourth gospel have been less extensive. The six author portraits are divided evenly between the gospels and the epistles. This means that the codex has lost one evangelist portrait, that of Matthew, and three author portraits from the praxapostolos section : Luke, John, and Paul. The frontispieces are located, one at the beginning of the manuscript (color plate), between the canon tables and text, and the other at the end of the codex (Fig. 15), where it once stood antecedent to a psalter, now lost. Both the frontispieces are full page pictures. Unfortunately, only one of the text illustrations, the Release of Peter from Prison (fol. ngr, Fig. 17), thus has an entire page to itself. However, one may be quite sure that the manuscript had several such full page illustrations in its original state. Indeed these would be the very first of the miniatured leaves to be cut out. Uniformly the text illustrations fill the width of the column of writing (10 cm.); but they vary considerably in height. The shortest is the Call of Matthew (fol. i6r, Fig. 10), which measures only 4.2 cm. in height. The tallest is the Anointing in Bethany (fol. 5 iv, Fig. 53), which measures 10.5 cm. Usually on miniatured pages approximately half the column is occupied by the illustration. Normally only one miniature appears on a given page. There are, however, two pages (ff. i6r and 86r, Figs. 10 and 1 1 ) that are enriched to the extent oftwo miniatures apiece. Almost without exception the miniatures immediately precede the texts they illustrate. B y this arrangement the mind is enabled to comprehend the miniature before

24

THE ART

BULLETIN

it does the text. A s a result the text serves as a commentary to the miniature, rather than the picture serving as an illustration for the text. This was a favorite arrangement with Byzantine miniaturists of the scriptures, for whom the sacred iconography as well as the sacred text equally conveyed the word of God. This, in fact, was almost the affirmation of that zealous defender of images, the Patriarch Nicephorus. 24 Most of the miniatures in Codex 2400 are single scenes. Nevertheless, there are two instances of continuous treatment which blend separate scenes into a single composition. In the baptismal miniature at the beginning of the first gospel (fol. i o v , Fig. 30) three distinct scenes arranged in two friezes are exhibited in vertical perspective, the action proceeding in each case from left to right. In the upper frieze at the left John the Baptist preaches repentance to the Pharisees and Sadducees; then on the right he enacts literally his affirmation regarding the Coming O n e : " A f t e r me there is coming one stronger than I am, one whose shoes I am not fit to stoop down and u n t i e " (Mark i : 7 not Matt, iii : 11). The cycle is completed in a contrasting scene in the frieze immediately below, where the baptism of Jesus by John is depicted. In the numbers of scenes and figures included this miniature is exceptional in Codex 2400. It reminds one of the great examples of continuous narrative in certain miniatures of the Vienna Genesis. On fol. 6 i r of the Chicago manuscript is another example, much more simple, of continuous style in narrative depiction. Again, these are scenes from the John the Baptist cycle, illustrating the summary statement in Luke iii: 18 if. about the Baptist's imprisonment (Fig. 44). The action proceeds from right to left in this instance. On the right the Baptist denounces Herod Antipas " b e c a u s e of Herodias, his brother's wife, and all the wicked things Herod had done." On the left John the Baptist is in prison as a result of his outspoken condemnation of Antipas. T h e continuous treatment of themes in miniatures always tempts one to infer an ultimate origin for the scenes in question in some ancient rotulus cycles. In the case of a manuscript so profusely miniatured as Codex 2400 this inference is particularly urgent. For typical Old Testament narrative cycles, the rotulus origin is made secure enough by such an impressive monument as the Joshua Rotulus. The equivalent cannot be said for N e w Testament narrative cycles. In the case of the Rockefeller McCormick miniatures in particular, the two instances of continuous representation just discussed are so altogether exceptional that they furnish no sufficient basis for inferring a rotulus origin, ultimate or more immediate, for the series as a whole. The ninety miniatures in the Chicago manuscript at once call for comparisons to make concrete the importance of the codex in the field of Byzantine iconography. It is difficult, however, to make even a quantitative comparison that will fairly indicate the eminence of this extensive series of miniatures. The statement has been made with strict accuracy that a more profusely illustrated Byzantine Testament is not known to exist. More might be said. Scholarship does not know today of another miniatured Greek N e w Testament that has text illustrations. In this particular Codex 2400 is simply unique. W h e n one investigates the B codices listed b y von Soden 25 one discovers that all the N e w Testaments declared to have Bilder are illus24. Antirrheticus, 100, Col. 382.

III, 5; Migne, Pair. Gr., Vol.

25. H. Freiherr von Soden, Die Schriften des Neuen Testaments I, 1, 102 ff.

C O D E X 2400 A N D I T S

MINIATURES

25

trated with author portraits and frontispieces only. This means that they average not more than ten or twelve miniatures apiece at most. Even when the New Testament is combined with psalter and canticles, as the Chicago Testament was originally, the two sections in combination do not present a total number of miniatures at all comparable to the number in Codex 2400 at present. The most extensively miniatured manuscript of this class on Mt. Athos today is Pantocrator 49, which contains only thirty-two miniatures altogether. In the New Testament section (which follows the canticles in Pantocrator 49) there are only twelve miniatures.® A t the Convent of St. Athanasius on Mt. Athos is Laura B 26, a combined New Testament and psalter written by the same scribe as Codex 2400. This also has only twelve miniatures in its New Testament section, and as many more distributed between the psalter and canticles.27 All such comparisons of the Rockefeller McCormick Testament with other manuscripts of its own class only serve to emphasize the uniqueness of the Chicago codex. Since, however, this manuscript is chiefly distinguished for its text illustrations, and these are mainly to be found in the gospels, it is pertinent to compare the codex with Byzantine illustrated gospels. The most extensively miniatured Greek gospels in existence at the present time are the well-known Grec 74 of the Bibliothèque Nationale and the Laurentian V I 23 of Florence, both of them dating from the eleventh to the twelfth century. According to M. Henri Omont, Conservateur des Manuscrits for the Bibliothèque, Paris 74 contains 372 miniatures. 28 According to M. Gabriel Millet, Laurentian VI 23 contains 288 miniatures. 29 In reality, a far larger number of scenes are represented in these codices than these numbers would suggest; for each miniature is a long, narrow frieze in which three or four different episodes are included. Actually, almost every scene in the gospel story is depicted in these very remarkable manuscripts. 30 It is perfectly obvious that Codex 2400 will not bear quantitative comparison with these great collections of miniatures, three or four times as large. But it does stand next after them among New Testament manuscripts in the number of its miniatures. To rank the Chicago manuscript as third most important for the extent of its contribution to our knowledge of New Testament iconography, is certainly as fair and perhaps as accurate an estimate as is possible. M. Gabriel Millet in his indispensible L'Iconographie de VÉvangile divides miniatured Byzantine gospels into two general groups: those of which the illustration is complete, or approximately so, Paris 74 and Laurentian VI 23 being the examples; and those gospel manuscripts that are only partially illustrated. 31 The gospels in Codex 2400 belong to the latter'group, and they stand at the very head of this partially illustrated class. Hence the Rockefeller McCormick miniatures are tableaux, 26. Spyridon P. Lambros, Catalogue of the Greek Manuscripts on Mount Athos, Vol. I (Cambridge, 1895), p. 98, No. 1083. 27. Spyridon and Eustratiades, Catalogue of the Greek Manuscripts in the Library of the Laura on Mount Athos (Cambridge, 1925), p. 15, No. 146 28. H. Omont, Evangiles avec Peintures Byzantines du XI« Siècle, p. 2.

29. A. Michel, Histoire de l'Art, p. 256.

Vol. 1, Part 1,

30. For a discussion of the Slavonic cognates of Paris Grec 74 see the brilliant article by Mile. Der Nersessian entitled, Two Slavonic Parallels of the Greek Tetraevangelia : Paris in The Art Bulletin, Vol. IX, No. 3, pp. 223-274. 31. G. Millet, Recherches sur l'Iconographie VEvangile (Paris, 1916), p. 8.

de

26

THE ART

BULLETIN

not friezes, which are found typically in the completely miniatured gospels. A n o t h e r eminent member of the partially illustrated group, the " F o u r Gospels of K a r a h i s s a r , " now L e n i n g r a d 1 0 5 , is iconographically so closely related to Codex 2400 as to be its " twin brother " — t o quote the pronouncement of P r o f e s s o r A . M. Friend, Jr., of Princeton. L e n i n g r a d 105 contains sixty-five miniatures. T h e Vienna tetraevangelia, Theol. G r . 1 5 4 , is credited with seventy-five miniatures. Y e t other noteworthy, but partially illustrated gospel manuscripts, are Berlin Q 66 with forty-two miniatures, Iveron 5 on Mt. Athos with its thirty-six miniatures, and the Graeco-Latin 54 of the Bibliothèque Nationale which exhibits thirty-one miniatures. In A m e r i c a the Pierpont Morgan L i b r a r y , of N e w Y o r k City, has a distinguished collection of miniatured Byzantines. A m o n g them the opulent Y a t e s T h o m p s o n L e c t i o n a r y includes thirtyfour miniatures, both marginal and tableaux. A l s o a magnificent cruciform lectionary in the Morgan collection, numbered 692, contains an equal number of scenes disposed in the open spaces at the four corners of the p a g e . T h e s e comparative statistics have been cited because they illustrate better than any other available data the extent of the iconographie contribution made b y the Chicago manuscript. P r o f e s s o r Millet has further emphasized that among the partially illustrated gospel manuscripts there are a few that illustrate the gospel of Matthew quite fully, and after that exhibit scenes that are peculiar to the other gospels, without repeating themes that have already appeared. The somewhat tattered G r e c 1 1 5 of the Bibliothèque Nationale is the classic instance of a manuscript which thus gives preference to the first gospel in its illustrations. A m o n g the versions Copte 1 3 , also of the Bibliothèque, is another prime example with an extensively illustrated Matthew section. Codex 2400, however, is not thus reticent about duplicating scenes from one g o s p e l book to another. T r u e , Matthew has more miniatures than any other book in the codex. B u t there are at least eight scenes in the Matthew section that may fairly be considered to have a second representation in some other gospel. These extra depictions, however, except in the case of the Transfiguration, are so very different from the Matthean tableaux as to be virtually new scenes. T h e y are really variant illustrations of parallel gospel narratives. A s such they present the " synoptic problem " in a fascinating pictorial form.

VI.

CANON

TABLES

AND

FRONTISPIECE

A preliminary gathering of four folios carries the canon tables of the Rockefeller McCormick g o s p e l s . T h e complete series of ten tables is given. T h e y are arranged in tall, slender arcades that support richly patterned façades (color plate). On top of the first canon table (fol. 2) is the chalice as the fountain of life with fronting peacocks, symbolizing the hope for immortality. Various interesting Byzantine birds; guinea hens, cranes, parrots, partridges, quails, and pigeons surmount subsequent tables. Foliated finials are prominent. A pair of lacertines with knotted necks terminate canon 5 (fol. 3v). T h e dominant colors in these designs are violet red and cobalt blue, with g o l d as a solid b a c k g r o u n d on the archivolt. T h e " A m m o n i a n " section numbers are written in blue. A s a series these canon arcades, with their g r a c i o u s

C O D E X 2400 A N D I T S

MINIATURES

27

dignity, formal regularity, and genuine richness, represent noteworthy phases of Byzantine art in its characteristic development. Of the numerous and varied miniatures in C o d e x 2400 the frontispiece inevitably commands first attention. Since it is done in gold and bright colors on a folio of purple vellum (fol. 6v, color plate), it is simply inescapable. Nothing could be any more characteristically Byzantine than the custom of writing in gold and miniaturing in colors on parchment leaves dyed a deep purple. This was a practice that appealed strongly to the sumptuous tastes of the Byzantine Greeks. Constantinople b e c a m e the g r e a t E a s t e r n center for the production of manuscripts written in this manner, with R o m e as a secondary center in the West. V e r y noteworthy examples of such lavish Byzantine book production are treasured in the great collections t o d a y : the Vienna Genesis (Theol. G r . 3 1 ) ; Codex Purpureus Petropolitanus (N), still in L e n i n g r a d ; R o s s a n e n s i s (S), in southern Italy, Sinopensis (0), in the Bibliothèque Nationale, and the G o s p e l s of B e l g r a d e (i>)—all dating from the sixth c e n t u r y — a n d the G r e e k Psalter at Zurich, dating in the seventh century. A related Byzantine custom was to inscribe or illuminate a single folio, a union of folios, or even a small gathering of purple leaves and insert them as a frontispiece or ornamentation into an otherwise normal manuscript. This custom developed in the eighth or ninth century after the Iconoclastic C o n t r o v e r s y , and was continued in the G r e e k E a s t particularly. A t the time of the R e n a i s s a n c e there was a t e m p o r a r y revival of this mode in the West. A t the convent of Vatopedi on Mt. Athos there is a tetraevangelia codex (Eustratiades N o . 949) which has its importance emphasized b y an inserted union of purple folios at the beginning, on which canon tables are illuminated in bright colors and gold. 132 A n even more striking example of the Byzantine custom of using a purple frontispiece to give distinction to a codex is found in the purple folio at the beginning of C o d e x 2400. When I showed a photograph of this p a g e to the late Mr. E . H . D r i n g of Quaritch C o m p a n y , the latter expressed the utmost surprise, and affirmed that in all his forty y e a r s of manuscript experience, not a single piece of purple vellum, least of all a miniatured piece, had ever passed through his hands. 3 3 It would be difficult to imagine a theme more fitting or impressive as a N e w Testament frontispiece than the depiction of Moses receiving the L a w on Mt. Sinai. Moses was the dominant prophet of the Old Testament. H i s law was the central item in the old covenant of J a h v e with his people. A l l this is s u g g e s t e d in the incomplete quotation from John i: 17 inscribed in letters of g o l d above the picture: " A s the law was given through Moses.... " With this as a preface and frontispiece, the whole N e w Testament that followed told the story of how " blessing and truth came to us through J e s u s C h r i s t , " and thus matched the new covenant with the old. In the frontispiece to C o d e x 2400 a long f a c e d but beardless Moses rushes f o r w a r d

32. Eustratiades and Arcadios, Catalogue of the Greek Manuscripts in the Library of the Monastery of Vatopedi on Mt. Athos (Cambridge, 1924), p. 174, No. 949. H. Brockhaus, Die Kunst in den Athos-Klostern (Leipzig, 1924 2 ), pp. 196, 199, 287. Brockhaus cites

this manuscript as No. 795. F . Perilla, Le Mont Athos (Paris, 1927), p. 138. 33. The House of Quaritch did not handle either Sinopensis (o) in 1899 or the Codex Purpureus of Leningrad (N) in 1896.

28

THE ART

BULLETIN

in a mountainous landscape to receive the law with veiled hands from the dextera domini extending to him from heaven in the upper right-hand corner of the picture. A single trefoil on the peak at the left suggests trees on Mt. Sinai ; the tree in the form of a leaf was a familiar Byzantine expedient from the tenth century on. The single figure of Moses, with its bold, simple modeling; its free pose, easy movement, and fluttering draperies; the painted peaks of the background and the blue of the sky above, all combine to make this frontispiece the most Alexandrian, illusionistic scene in the Rockefeller McCormick manuscript today. It is distinctly Hellenistic in spirit and manner. Moses receiving the law is a very familiar figure in Byzantine miniatured manuscripts. The primitive type was probably a single standing frontal figure with scroll in hand or large tablet of the law supported on his arm—a figure modeled after the standing orator of the classical period. It is thus that Moses appears among the Old Testament worthies in the Peshitto Old Testament of Paris (Bibliothèque Nationale, Syr. 341) dating in the seventh century. 31 In the sixth century Syriac gospels of Rabula in Florence (Laur. I 56, 3v) Moses is depicted as a single figure facing front, but reaching up to receive the law from the hand of the Eternal, much as he does in the Chicago frontispiece. 35 A distinct development was the depiction of Moses receiving the law as an incident in a full scene with a great variety of appropriate accessories. This became, in fact, a favorite composition in Septuagint Old Testaments, in illustrated octateuchs, in psalters both of the aristocratic and the monastic-theological types, and in manuscripts of Cosmas Indicopleustes. Moses receiving the law as a frontispiece scene was probably a conflation of the simple, restrained portrait type, and the elaborate, full scene of Septuagint manuscripts. In the frontispiece the Moses portrait was worked out in greater detail, and the full scene of receiving the law was markedly simplified. Direct parallels to the frontispiece Moses of Codex 2400 are well known. In a ninth century uncial lectionary of the gospels in the Vatican collection (Gr. 1522) there is a very vigorous rendition of the scene. B y contrast, in the John Rylands Library of Manchester there is a very restrained, kneeling Moses who receives the law in a gospels manuscript of the twelfth century (Gr. 17). The closest cognate of all, however, is the frontispiece Moses of Suppl. Gr. 1 3 3 5 in the Bibliothèque Nationale (Fig. 12), a combined New Testament and psalter, otherwise closely related to Codex 2400. This Bibliothèque Moses is portrayed on an inserted union of parchment folios completely different in color and quality from the rest of the manuscript. The color is a curious old-rose shade, somewhere between purple and old ivory. It is possible that the union was originally dyed purple, and that it has faded to this strange color. If so, the parallelism with the Chicago frontispiece becomes surprisingly close.

34. H. Omont, Peintures de VAncien Testament dans un Manuscrit Syriaque (Monuments et Memories, Fondation Euglne Piol, t. 1 7 , Paris, 1909), pl. I V , No. 4.

35. S . E . Assemanus, Bibtiothecae Mediceae Laurentiae et Patatinae Codicum Manuscriptorum Catatogus (Florence, 1742), pp. 1 if., pl. 4.

CODEX

2400 A N D I T S

V I I . AUTHOR

MINIATURES

29

PORTRAITS

T h e g o s p e l frontispieces in C o d e x 2400, which are evangelist portraits, constitute an independent c y c l e of singular coherence and interest. T h e y stand together in format, in decorative quality, in figure treatment, and in design. Unfortunately, only three of the evangelists remain: Mark, L u k e , and John (color plate and F i g s . 8, 14). Of the three portraits, that of the fourth evangelist is perfectly square, and the other two are nearly so. This is an unusual format for evangelist portraits, which are more frequently oblong on a vertical axis, the shape of the picture conforming to the shape of the p a g e of the a v e r a g e cursive manuscript. T h e square format, on the other hand, recalls the nearly square shape of the great early uncial manuscripts, like Vaticanus (B) or Sinaiticus (x), written three and four columns to the page. In conformation to such pages, wide and square, the full-page miniatures of the early uncials were square in shape and sometimes widely framed, like the sequence at the beginning of the V i e n n a Dioscurides (Med. G r . i). 36 This primitive format is recalled b y the square shape of the evangelist portraits in Codex 2400. T h e wide, intricate, formal, patterned borders, lavish with gold and heavy with color, that frame these portraits, are another significant feature of the g o s p e l headpieces. Such opulent frames are to be found only in that masterpiece of middle Byzantine style, Urbinate Greek 2 of the Vatican, 8 7 which was executed for John Comnenus in 1 1 2 8 - 9 . These are not Hellenistic borders with chastened classic patterns, such as one finds on the p a g e s of the famous Paris Psalter (Grec 139). 3 8 T h e y are, rather, Oriental decorative work, consisting of geometrical repeat patterns and conventional flower and leaf designs, done f o r the most part in blues and reds and greens, with relieving touches of white and groundwork of gold. Persian carpets or enamels furnish the best parallels to such richness and beauty as here displayed. This is middle B y z a n t i n e ornament at its best. Within these enframements the evangelists are seated in profile, each with his face turned slightly toward the front (color plate and F i g s . 8, 14). E a c h evangelist is seated on a fat cushion that eases a decorated b e n c h ; and each has his feet disposed on a l a r g e footstool that is c o n s p i c u o u s l y pearled on the sides. A writing desk and a lectern complete the furniture. T h e b a c k g r o u n d is a solid panel of gold save in the case of Mark, who has back of him an elaborate building with a l o g g i a in its upper story and a conical tholos on its roof (color plate). In the matter of p h y s i o g n o m y the evangelists of Codex 2400 correspond in general to the s u m m a r y characterizations of Byzantine evangelists given b y Mr. J . A . Herbert in his Illuminated Manuscripts: " S t . Matthew is always ¿n old man, with white hair

36. A . von Premerstein, C . W e s s e l y , I. Mantuani, Dioscurides, Codex Aniciae Iulianae picturis illustratus (Codices Graeci et Latini, T o m . X , L e y d e n , 1906). Facsimile. 37. C. Stornajolo, Miniature delle Omilie di Giacomo Monaco e del? Evangeliario Greco Urbinate (Codices e Vaticanis selecti, Series Minor, V o i . 1 , R o m e , 1910). Facsimile.

38. H . Omont, Miniatures des Plus Anciens Manuscrits Grecs de la Bibliothèque Nationale (Paris, 1929), pp. 1 - 1 0 , pl. I - X I V bis. Recall particularly Professor Morey's dating of the Paris Psalter by the J o s h u a Rotulus as of the seventh century: Notes on East Christian Miniatures, in The Art Bulletin, Vol. X I , N o . I (March 1929), pp. 21-50.

THE ART

3°

BULLETIN

and beard. St. Mark is much younger, dark haired, sometimes of a strongly Semitic type. ... St. L u k e is a y o u n g man in his prime, fair, with g o o d features of Greek type, and slightly pointed b e a r d ; sometimes tonsured St. John is generally depicted as an old man, with long white beard and bald head, the forehead v e r y large and dome-shaped." 3 9 In significant details of posture the evangelists of Codex 2400 repeat poses that are familiar in tenth century cycles, from the period of the so-called Macedonian Renaissance. With no single manuscript are the relationships more exact than with a gospels codex in the Athens Bibliothèque numbered 56, written in the tenth century by a monk named John 4 0 ( F i g s . 18-20). T h e Mark of the Chicago codex (color plate) is the Matthew of the Athens gospels : chin on left hand, right hand with pen resting on codex in evangelist's l a p ; right leg diagonal in relation to the vertical left leg. Similarly, the L u k e of the Chicago codex ( F i g . 1 4 ) — e x c e p t for his b e a r d — i s the Athens L u k e who bends forward to dip his pen into an inkwell. Most astonishing and significant of all is the correspondence between the two J o h n s in these codices ( F i g . 8). B o t h are venerable, stoop-shouldered figures who bend heavily over a ponderous, half-opened tome held in the lap. These comparisons make it plain that the evangelist c y c l e in Codex 2400 g o e s back at least to the neo-classical revival of the tenth century. In individual instances a m o n g the evangelists an even earlier genesis can be traced. P r o f e s s o r A . M. Friend, J r . , in his masterly study of evangelist portraits in Greek and Latin manuscripts, has succeeded in isolating six really primitive and fundamental types among the variety of seated evangelists of the tenth century. 4 1 F o u r of the six he finds in Vatican G r . 3 6 4 , an eleventh century manuscript, highly refined in style, with the particular refinement of incised architectural settings in the gold of the backgrounds. T h e other two seated evangelist types are represented b y L u k e and John in Stauronikita 43, a tenth century codex on Mt. A t h o s , the most weighty of Byzantine manuscripts in the matter of the portraiture of seated evangelists. Comparing the Chicago series with the evangelists of Vatican G r . 364 the similarities are found to be remarkably close, almost as close, in fact, as with Athens 56. In the portrayal of Mark, particularly, the resemblance approaches the point of identification ( F i g . 21). Indubitably, then, the Mark of Codex 2400 is a primitive type. Comparison proves that the same may be affirmed of the portrait of L u k e . When the third evangelist is placed alongside the youthful L u k e of Stauronikita 43 ( F i g . 22), the two are seen to be v e r y like each other; just as, indeed, the L u k e of Athens 56 exactly reproduces the L u k e of Stauronikita 43, without the stage scenery. That very theater scenery, however, plus the posture of the figure, proves the type to be primitive. 39. J . A . Herbert, Illuminated Manuscripts (London, 1912 2 ), p. 64. F o r exact illustrations of the Herbert description see the evangelists in the Rixhy Gospels now in the Henry E . Huntington Library at San Marino, California. T h e y are reproduced in part in Professor E d g a r J. Goodspeed's The Bixby Gospels (Chicago, 1915). 40.

j . Sakkelion, Catalog

of Manuscripts

of

the

National Library of Hellas (Athens, 1892), p. 1 1 . P. Buberl, Die Miniaturen Handschriften der Nalionalbibliolhekin Athen (Vienna, 1917), pp. 7 ff., pis. V I and V I I . 4 1 . A . M. Friend, J r . , The Portraits of Evangelists in Greek and Latin Manuscripts, in Art Studies, Vol. V (Cambridge, 1927), pp. 134 ff.

FIG. 14—Chicago, Rockefeller McCormick New Testament. Evangelist

FIG. 15 — Chicago, Rockefeller Mc Cormick Collection: Byzantine New Testament. David the Musician

Collection: Luke

Byzantine

FIG. 16—Paris, Bibl. Nat.: Suppl. Gr. 1335. David the Musician

Fir.. I 7— Chicago, Rockefeller McCormick Collection : Byzantine New Testament. Peter's Release front Prison

FIG. IG—Evangelist Eike

(After

Buberl)

Athens,

National

Fui, IH—Athens, National Library : Ms. ¡6. Evangelist Matthew (After Buberl)

FIG. 20—Evangelist John (After Library:

Ms.

56

Buberl)

CODEX

2400 A N D I T S

MINIATURES

33

T h e distinctive evangelist in Codex 2400 is John. He is not the John of Stauronikita 4 3 nor yet the John of Vatican Gr. 364, though he resembles the latter somewhat. Emphatically, he is not the standing John who dictates to Prochorus in middle Byzantine manuscripts from the eleventh century on. This amusing and immensely popular figure of the middle Byzantine period, who monopolized the scene for a time, is definitely repudiated by our portrait artist in favor of the more dignified, more classical fourth evangelist of the Macedonian Renaissance who is immortalized in Athens 56. Just here is the outstanding and permanent contribution of Codex 2400 to the study of evangelist portraiture. Without exception the Chicago gospels and a large group of cognate manuscripts—Coislin 200, Laura B 26, Suppl. Gr. 1 3 3 5 , Leningrad 1 0 5 , and others—carry this revived tenth century figure as the portrait of the fourth evangelist. In place of the middle Byzantine John-Prochorus combination an earlier classical revival—John with a half-opened tome—was definitely substituted. This became the typical fourth evangelist portrait of the Paleologan period. I t is well known that Professor Friend has carried the origins of his seated evangelist types far back of the tenth century, to E a r l y Christian times ; and he has localized some of these types in Ephesus, the great center of art and literature for western Asia Minor. 4 2 Moreover, by a series of comparisons of seated evangelist portraits with Hellenistic statues of poets and philosophers, and further comparisons of architectural portrait backgrounds with Hellenistic stage fronts, he has demonstrated the ultimate origins of the seated evangelist types in the honorific statues of philosophers and poets with which the Greeks and R o m a n s used to decorate the façades of their stage buildings. W i t h these Hellenistic origins the evangelists of Codex 2400 connect through the neo-classical revival of the tenth century. T h r e e author portraits antecedent to epistles in the Chicago New Testament constitute another distinct and independent portrait sequence of a very high order. Most unfortunately, the cycle is broken : James, Peter, and J u d e remain; but John and Paul have been lost. W h e n photographs of the manuscript were shown to Mrs. McCormick for the first time, these portraits were the only miniatures she took time to examine. On the basis of these three portraits alone she decided to purchase the manuscript. Certainly they are among the most impressive of the miniatures in Codex 2400. T h e three authors are shown as standing half figures, each with a scroll or codex in his left hand and with his right hand raised in a speaking gesture of benediction. James is depicted in a more formal, exceptional manner than the other two, as a patriarch of the Eastern church, wearing a broad omophorion with crosses embroidered thereon, a sleeved sticharion and the phelonion (Fig. 56). In his hand he carries a codex, its covers set with semiprecious stones. His head is haloed, but unmitered. T h e background of the portrait is a panel of plain gold cut off at the bottom by dark-green ground-work. T h e inscription reads 0 AriOS IAKQB02 AAEAH>02 eE[Or?] " St. James, Brother of God. " If our reconstruction of the ending of the inscription is c o r r e c t — a n d all the remaining lines are in its favor—then high Christology is 42.

Op cit., pp. 140 ff.

34

THE ART

BULLETIN

here recorded. Deity is here ascribed to Jesus Christ through the medium of His brother James. Available indexes do not disclose another instance of this striking titulus for James the Righteous of Jerusalem. Apparently Codex 2400 here preserves a unique epigraphic record of lofty Eastern Christology. The other two portraits, of Peter and Jude, group together (Figs. 55, 57). Both authors are dressed in secular costume. Peter has black clavi and Jude has red clavi. Both carry scrolls, instead of codices. With a refined sense for harmony the scroll in each case is wound with narrow bands that correspond in color to the clavi on the author's garments. In each case the author's figure is flanked by architectural constructions that fill in the space admirably. The dark groundwork is cut from the bottom of Peter's portrait by the beginning of the first epistle. In both cases the inscription simply names the saint. The Petrine features are unforgettable; a roundish face, thick clustering hair about the forehead, and a close-cropped beard. B y contrast, the features of Jude are much elongated and his beard pointed, though split at the end. In fact, the features of Jude are almost exactly the features of his older brother James, the one difference being that Jude's hair and beard are nut brown, while the hair and beard of James are black turned to gray. This portrait sequence relates immediately and specifically to the praxapostolos author cycle consisting of Luke, as the author of Acts, and James, Peter, John, Jude and Paul as writers of epistles—six authors altogether. As a praxapostolos cycle the Chicago portraits are true to type; for usually the authors in this section of the New Testament were portrayed as standing rather than seated figures. In an eleventh century praxapostolos of the Vatican (Gr. 1208), which is said to have belonged to Carlotta, Queen of Jerusalem, by whom it was presented to Pope Innocent III, these six authors stand at the very beginning of the codex, arranged in three pairs according to literary sequence. Quite exceptional is the tenth century Bodleian praxapostolos (Canon. Gr. 1 1 0 ) in which the five authors (James is missing) are shown seated, in the manner of the Ephesian evangelists. Of all the praxapostolos authors Luke is the only one who shows any pronounced tendency to remain seated. In his case this posture is explicable as due to the influence of the seated evangelist Luke. Even when he is thus pictured at the beginning of Acts, however, he is invariably differentiated in marked ways from the preceding evangelist Luke. Leningrad 1 0 1 , a twelfth century Byzantine Testament, textually related to Codex 2400 in I Corinthians, has a seated Luke at Acts, connecting a series of seated evangelists with a sequence of standing apostles. The Luke of the third gospel, however, is taking dictation from Paul, while the Luke at Acts is seated alone copying from a scroll. With the exception of Luke, praxapostolos authors almost invariably stand, as do the Chicago trio. Standing apostles in the second main section of the New Testament demand for consistency standing evangelists in the first main section. This type has become very familiar to students of gospel iconography as one of the two main evangelist groups to be found in Greek manuscripts—thanks to the indefatigable labors of Professor Friend, who has traced the historical development and the ultimate genesis of the

C O D E X 2400 A N D I T S

MINIATURES

35

standing evangelists. 4 3 A c c o r d i n g l y , in such great tetraevangelia as the Nazarin Gospels of the Bibliothèque Nationale (Grec 70), the Gospels of St. Catherine's on Mt. Sinai (Ms. 204), and Andreaskita 5 (now in an American collection) one may find the evangelic counterparts for the standing authors in Codex 2400. N o r should one forget the unique painted book-covers belonging to the Freer Gospels in Washington (W) ; for these attest the honest attempt of Coptic E g y p t of the seventh century to supply standing evangelist portraits for an important but unillustrated gospels manuscript. 44 B a c k of the standing apostles and evangelists of the N e w Testament are the standing prophets of the Old Testament, best known, perhaps, from the three Roman manuscripts published together by Munoz: Chigi R VIII 54, of the eleventh c e n t u r y ; Vatican Gr. 755, of the tenth century; and Vatican Gr. 1153, of the twelfth century. 45 Quite as noble, however, is the prophet of the Laurentian manuscript (Plut. V 9) which dates from the same period. 46 In comparing these prophet sequences with the standing author portraits in the N e w Testament it is particularly noticeable that the prophets uniformly carry scrolls, while the New Testament authors are more frequently equipped with books. A few outstanding monuments bring together these compact author cycles and, with the addition of y e t other figures, give a clear intimation of what the sequence was in its fullest extent. In the sixth century Rabula Gospels of Florence (Laur. I 56)" with its singular combination of N e w Testament scenes and Old Testament characters, Moses and A a r o n , Joshua and Samuel, David and Solomon, all precede the alignment of prophets, major and minor. W i t h the exception of Solomon, whose kingly dignity demands that he be seated, and Jonah, who sleeps under his gourd vine, these are all standing figures. Then follow two seated evangelists and two standing evangelists. This juxtaposition of the Old Testament worthies and the N e w Testament worthies is a distinctive feature of this remarkable codex. A g a i n , it is observable that the prophets generally carry scrolls, while the evangelists, three out of four, have books. 48 A more inclusive, more officialized, less literary sequence is found in the single figure mosaics of S. Apollinare N u o v o in Ravenna, that alternate with the windows, below the range of gospel scenes and above the great processions. 4 9 T h e y date from the sixth century. The figures are unlabeled, either by names or by symbols. Most of them carry scrolls; but a few have codices. T h e y number thirty-two, and are obviously intended to represent the four major prophets and the twelve minor

43.

Op.

cit.,

pp.

124-133.

44. C . R . M o r e y , East Christian Paintings in the Freer Collection ( N e w Y o r k , 1914), p p . 63-81. 45. A . M u ñ o z , / Codici Greci Miniati delle Minori Biblioteche di Roma ( F l o r e n c e , 1905), p p . 13-44. 46. A . M. B a n d i n i u s , Catalogus Codicum scriptorum Bibliothecae Mediceae Laurentianae r e n c e , 1764), p p . 19-22, p i .

II.

47. R . G a r r u c c i , Storia deWArle (Prato,

Manu(Fio

Cristiana, V o l . I l i

1876), p l s . 128-40.

5 . E . A s s e m a n u s , Bibliothecae

Mediceae

Lavrentiae

el Palatinae Codicum Manuscriptorum Catalogus r e n c e , 1742), p i s . I - X X V I , p p . r-25.

(Flo-

48. R . G a r r u c c i , op. cit., p i s . 135, 136; S . E . A s s e m a n u s , op. cit., p i s . X V I , X V I I . 49. O . M. D a l t o n , Byzantine Art and Archeology ( O x f o r d , 1911), p p . 350-355. A s i m i l a r s e r i e s h a s b e e n r e c o n s t r u c t e d b y H e l e n W o o d r u f f in t h e c a s e of t h e lost m o s a i c s of L a D a u r a d e a t T o u l o u s e . S e e The Art Bulletin, V o l . X I I I , N o . 1, p p . 80-104. C . Ricci, Ravenna ( B e r g a m o , n . d . 10), p i s . 55-86. B e r c h e m a n d C l o u z o t , Mosaiques chrétiennes (Paris,

1924), p p .

125-141.

36

THE ART

BULLETIN

prophets in the Old Testament, with the corresponding four evangelists and twelve apostles of the N e w Testament. Of all the standing figure c y c l e s examined thus far, these mosaics most resemble the author portraits of Codex 2400 in monumental dignity, figure proportion, facial expression, treatment of hair, etc. These comprehensive series show that the C h i c a g o author portraits belong to a grand cycle of Old and N e w Testament p e r s o n a g e s , the authors of or the leading characters in the combined J e w i s h and Christian scriptures. In the full cycle the author preferably was portrayed. If he was not known, a character of the book might be pictured instead. T h e great Peshitto Old Testament, of the seventh century, now in the Bibliothèque Nationale (Syr. 341), 5 0 gives an excellent idea of how a G r e e k Bible thus illustrated with author portraits must have looked. A t the beginning of almost every book, painted against a neutral b a c k g r o u n d and at the head of a column of writing, a supposed author stands and gesticulates. M. Omont is of the opinion that all the portraits in this S y r i a c manuscript were copied from Greek models. T h e series extended through the N e w T e s t a m e n t ; for in its original state the Paris manuscript was a complete Bible. F r o m such a vast series the portraits in Codex 2400 were derived, distantly though not immediately perhaps. P r o f e s s o r F r i e n d ' s researches in evangelist origins have demonstrated the initial emergence of the standing author type in Hellenistic E g y p t . 0 1 T h e codices carried b y N e w Testament authors and the rolls carried b y Old Testament prophets indicate that the former series e m e r g e d during the first Christian centuries when the codex was coming into vogue, and that the latter series developed during the last centuries of the pagan e r a — t h e period of the S e p t u a g i n t — w h e n the p a p y r u s roll was the standard book form. In this matter the author portraits in C o d e x 2400 show mixture; one apostle carries a codex and the other two have rolls. This would be expected as late as the thirteenth century. T h e pagan prototype f o r the standing author, Jewish or Christian, can easily be recognized. Characteristic features of the type are these : a fully robed masculine figure in statuesque pose, a scroll held in the left hand, the right hand raised in a speaking gesture. In this manner Greek and R o m a n artists characteristically portrayed their great orators. 52 A s M. Ebersolt s u g g e s t e d in the case of the standing evangelists, 5 3 and as Mr. Dalton earlier suggested in the case of the standing prophets, 64 both alike g o back ultimately to the standing orator statues of classical antiquity. H e r e is the final derivation of the trio of epistle authors in Codex 2400. In their earliest literary use the standing authors were painted against the plain background of the papyrus roll itself, as the prophets stand portrayed in the late fifth century fragments of an Alexandrian world chronicle in the Golenishev collection. 55

50. H. Omont, Peintures de VAncien Testament dans un manuscrit syriaque (Monument et Mémoires, Fondation Eugène Piot, t. 17, Paris, 1909), pp. 85-98. pis. V - X . 51. A . M. Friend, J r . , The Portraits of the Evangelists in Greek and Latin Manuscripts, in Art Studies, Vol. V (Cambridge, 1927), pp. 126 ff. 52. Salomon Reinach, Repertoire de la Statuaire Grecque et Romaine, 5 vols. (Paris, 1897-1924),passim.

53. J . Ebersolt, La Miniature Byzantine (Paris and Brussels, 1926), p. 82, notice 32. 54. O. M. Dalton, Byzantine (Oxford, 1 9 1 1 ) , p. 473.

Art and

Archeology

55. Bauer-Strzygowski, Eine alexandrinische Weltchronik (Denkschriften der kaiserlichen Akademie der Wissenschaften, Phil. hist. Klasse, Bd. 5 1 , Vienna, 1906),

FIG. i G—Slaitghier of Bethlehem Children

FIG. 25—Nativity of Jesus

FIG, 24 —John the Baptist Preaching the Kingdom

FTS. 26 —Hie

Beatitudes

FIG. 27—Jesus' Denunciation of

FIG. 28—Grain Picking 011 the

Galilean Cities

Sabbath

Chicago, Rockefeller McCormick

Collection: Byzantine New Testament

C O D E X 2400 A N D I T S

MINIATURES

39

Thus they appear also against plain parchment, each within a column of writing, in the ninth century Vatican codex of Cosmas Indicopleustes (Gr. 6g9).68 B y comparison with the primitive simplicity and directness of this earliest manner of portrayal, the elaborate settings and frames provided for the authors in Codex 2400 are seen to be highly artificial (Figs. 55-57). Standing figures that naturally require vertical space are cut off and crowded into horizontal frames. To fill up the spaces at the sides architectural constructions are introduced. These thirteenth century additions of frames and architecture are persuasively rendered; but they do not belong to standing author portraits. A collocation of these portraits with corresponding cycles in related thirteenth century manuscripts shows that the mode of the day was to fit the standing author figure into an elaborate setting of some sort. The artist who painted the standing authors in Suppl. Gr. 1 3 3 5 left three of them full height, but filled in the backgrounds with buildings, and in one instance added a foreground landscape. 57 The painter of Paris 61 portrayed his Apostle Paul standing in an arched doorway flanked with panels of grill work. 58 A greater sense of propriety was shown by the portraitist of Coislin 200 when he left his backgrounds neutral and contented himself with reducing his standing authors to half figures only 59 (Figs. 58-60). A favorite expedient of the times was to crowd the standing author into a small medallion which was placed in the center of a large rectangular frame.60 Along with these varied contemporary adaptations the portraits in Codex 2400 exhibit the attempt—sincere if ill advised— to provide an ornate setting for the picture of a revered author. VIII.

TEXT

ILLUSTRATIONS

In an introductory monograph on a manuscript that contains eighty-two text illustrations it is obvious that detailed discussion cannot be given to individual miniatures. The best one can do is to treat this huge number of pictures in cycles and to isolate single miniatures only when they are typical of the manuscript or exceptional among manuscript illustrations generally. However, it is very useful to students of iconography to have at least a complete list of subjects and texts illustrated. The titles and texts of the miniatures in Codex 2400 are accordingly listed as follows: Fol. 6v, Frontispiece: Moses Receiving the Law, John i : 17. gr, The Nativity of Jesus, Matt, ii : 1, 2. gv, The Slaughter of Boys in Bethlehem, Matt, i i : 16-18. 1 or, John the Baptist Preaching the Kingdom, Matt, iii: 1-9. 56. Monsignor Cosimo Stornatolo, Le Miniature della Topografia Cristiana di Cosma Indicopleuste ( Codices e Vaticanis selecti X, Rome, 1908). Facsimile. 57. Paris, Bibliothèque Nationale, Suppl. Gr. 1335, fols. 13IV, 169V, 184V. 58. Paris, Bibliothèque National, Grec 61, fol. 2i6r. Cf. also the author portraits in the epistles section of Leningrad 101 {twelfth century) on the following folios: 192, 196, 202, 207, 209.

59. Paris, Bibliothèque Nationale, Coislin 200, fols. l88r,

192V, 202r, 207r,

2ior.

60. Athos, Laura B 26, fols. i4or and I49r. Further references are not given since my file of photographs from this manuscript is incomplete. Berlin, State Library, Greek Octavo 13, fols. I27r, '3or,

I35r,

138V.

Paris, Bibliothèque Nationale, Grec 6 1 , fols. I98r, 209V,

214V.

THE

ART

BULLETIN

i o v , J o h n ' s W a r n i n g to P h a r i s e e s and S a d d u c e e s ; H i s S u b s e r v i e n c e to the C o m i n g O n e ; H i s B a p t i s m of J e s u s , M a t t , i i i : i o-1 7. 1 i v , The Beatitudes, Matt, v : 1 - 1 2 . I 4 r , J e s u s C u r e s a L e p e r , Matt, v i i i : 1-4. 14V, J e s u s a n d the C a p e r n a u m C e n t u r i o n , Matt, v i i i : 5 - 1 3 . i 5 r , T h e H e a l i n g of P e t e r ' s M o t h e r - i n - L a w , Matt, v i i i : 1 4 , 1 5 . 15V, J e s u s E x o r c i s e s T w o G a d a r e n e D e m o n i a c s , Matt, v i i i : 2 8 - 3 2 . i 6 r , T h e C u r e of the C a p e r n a u m P a r a l y t i c , Matt, i x : 2-8. i 6 r , T h e C a l l of M a t t h e w , Matt. i x : g . i 6 v , A n O f f i c i a l A p p e a l s to J e s u s , Matt, i x : 1 8 , 1 9 . i 8 r , J e s u s ' D e n u n c i a t i o n of G a l i l e a n Cities, Matt, x i : 20-24. i 8 v , G r a i n P i c k i n g on the S a b b a t h , Matt, x i i : 1-8. igr, J e s u s R e s t o r e s a W i t h e r e d H a n d , M a t t , x i i : 9 - 1 3 . 19V,

J e s u s C u r e s a B l i n d - a n d - D u m b D e m o n i a c , Matt, x i i :

22-24.

20V, J e s u s ' T e a c h i n g a b o u t T r u e K i n s h i p , Matt, xii: 46-50. 2 2r, S a l o m e a n d the H e a d of J o h n the B a p t i s t , Matt, xiv : 6 - 1 1 . 2 2V, T h e F e e d i n g of the F i v e T h o u s a n d Men, M a t t , x i v : 1 5 - 2 1 . 24V, T h e T r a n s f i g u r a t i o n , Matt, x v i i : 1-6. 3 i r , T h e L a s t S u p p e r , Matt, x x v i : 20-25. 3 2 r , T h e B e t r a y a l , Matt, x x v i : 4 7 - 5 1 . 33V, T h e C r u c i f i x i o n , Matt, x x v i i : 3 5 - 3 7 . 34V, T h e T w o M a r y s A d o r e the R i s e n J e s u s , Matt. x x v i i i : 8 - i o . 36r, The Evangelist Mark, Mark i : 1-3. 36V, J o h n the B a p t i s t P r e a c h i n g R e p e n t a n c e , M a r k i : 4-8. 3 7 r , J e s u s E n d o w e d with the S p i r i t , M a r k i: 1 0 - 1 2 . 37V, J e s u s E x o r c i s e s a C a p e r n a u m D e m o n i a c , M a r k i: 2 1 - 2 8 . 38r, J e s u s Cures Various Diseases, Mark i : 32-34. 38V, T h e C u r e of the C a p e r n a u m P a r a l y t i c , M a r k i i : 1 - 1 2 . 39V, J e s u s R e s t o r e s a W i t h e r e d H a n d , M a r k i i i : 1-5. 4or, J e s u s ' T e a c h i n g a b o u t T r u e K i n s h i p , M a r k iii: 3 1 - 3 5 . 40V, J e s u s T e a c h i n g f r o m a B o a t , M a r k i v : 1-9. 4 1 V , J e s u s Quiets the S t o r m , M a r k i v : 3 5 - 4 1 . 4 2 r , T h e M a n with a L e g i o n of D e m o n s , M a r k v : 1 - 1 3 . 42V, J a i r u s A p p e a l s to J e s u s , M a r k v : 2 2 , 2 3 . 4 3 r , T h e W o m a n with the H e m o r r h a g e T o u c h e s J e s u s ' G a r m e n t , M a r k v : 24-34. 43V,

Jesus Raises Jairus' Daughter, Mark v :

37-42.

46V, T h e J o u r n e y to the M o u n t of T r a n s f i g u r a t i o n , M a r k i x : 2. 5 i v , T h e A n o i n t i n g in the H o u s e of S i m o n the L e p e r , M a r k x i v : 3-9. 54V, T h e W o m e n at the T o m b , M a r k x v i : 1 - 7 . 5 6 r , T h e E v a n g e l i s t L u k e , L u k e i : 1-4. 56V, T h e A n n u n c i a t i o n to Z e c h a r i a h , L u k e i : 5-2o. 5 7 r , T h e A n n u n c i a t i o n to M a r y , L u k e i : 2 6 - 3 8 . 57V, T h e V i s i t a t i o n , L u k e i : 3 9 - 5 5 . 5 8 r , T h e N a t i v i t y of J o h n the B a p t i s t , L u k e i : 5 7 , 58. 5 g r , T h e N a t i v i t y of J e s u s , L u k e i i : 6 - 1 4 .

C O D E X 2400 A N D I T S

MINIATURES

4

59V, 6ir, 6iv, 62 r,

The Presentation, L u k e i i : 22-32. John the Baptist R e p r o v i n g H e r o d ; John in Prison, L u k e iii : 19, 20. Jesus Endowed with the Spirit while Praying, L u k e iii : 2 1 , 22. The First Temptation, L u k e iv : 1-4. 6 2 V , Jesus Reading the Lesson in the Nazareth S y n a g o g u e , L u k e i v : 1 6 - 1 9 . &3r, J e s u s Teaching in the Nazareth S y n a g o g u e , L u k e iv : 20-27. 6 3 V , Jesus Exorcises a Capernaum Demoniac, L u k e iv : 3 1 - 3 7 . 65V, The Raising of the Widow's Son at Nain, Luke vii : 1 1 - 1 7 . 6 6 v , The Anointing in the House of Simon the Pharisee, Luke vii : 3 6 - 3 9 . 68v, The Mission of the Twelve Disciples, L u k e ix : 1-6. 6 9 v , The Transfiguration, Luke i x : 2 8 - 3 5 . 77V, J e s u s Cures Ten Lepers, Luke x v i i : 1 1 - 1 4 . 84V, Jesus Blesses His Disciples, Luke xxiv : 50, 5 1 . 85r, The Evangelist John, John i : 1. 85V, The Descent to Hades, John i : 6 - 1 3 . 86r, The Testimony of John the Baptist to the Pre-existent One, John i : 1 5 . 86r, Moses Receiving the L a w , John 1 : 1 7 . 86v, Jesus Reveals G o d , John i : 18. 87r, The Testimony of John the Baptist to the L a m b of God, John, i : 2 9 - 3 1 . 87V, The Wedding Feast at Cana, John ii : 1-4. 88v, A R o y a l Officer A p p e a l s to Jesus at Cana, John iv : 46-50. 8gr, Jesus Cures a L a m e Man at Bethesda, John v : 1-9. 90V, The Feeding of Five Thousand Men, John vi : 3-14. g6r, The Raising of Lazarus, John xi : 38-44. 98r, Jesus Washing the Disciples' Feet, John xiii : 4 - 1 1 . io6r, Peter and the Jerusalem Leaders, Acts i : 15-22. io6v, The Election of Matthias, Acts i : 23-26. i o 7 r , Pentecost, Acts i i : 1-4. io8v, The L a m e Man at the Beautiful Gate, A c t s i i i : 1-8. i o g v , Peter and John before the Sanhedrin, Acts iv : 5 - 1 3 . 1 1 ir, The Death of Ananias, Acts v : 1-5. 114V, The Stoning of Stephen, Acts v i i : 55-60. 1 i 5 r , Paul L e d to Damascus, Acts ix : 3-9. 1 1 5 V , The Baptism of Paul, Acts ix : 1 7 , 18. 1 1 6 v , Peter Restores Dorcas to Life, Acts ix : 36-41. 1 i 7 r , Peter's Vision, A c t s x : 9 - 1 6 . 119V, Peter's Release from Prison, Acts xii : 5-9. 1 2 2 v , Paul Cures a L a m e Man at Lystra, A c t s xiv : 8 - 1 1 . i 3 8 r , The Apostle James, James i : 1. i 4 i r , The Apostle Peter, I Peter i : 1, 2. i 5 o r . The Apostle Jude, Jude 1, 2. 2o6v, K i n g David the Psalmist, I Sam. x v i : 18. Unfortunately, there are no extended primitive cycles of gospel illustrations with which to compare the miniatures of Codex 2400. That there were such cycles dating

42

THE ART

BULLETIN

to the fifth if not even to the fourth century cannot be doubted. The Codex of Rossano (2) and the Gospels of Rabula in Florence (Plut. I 56) give fugitive glimpses of the quality of these early illustrations. Also the literary attestation of such primitive sequences is ample and excellent. The Patriarch Nicephorus, outstanding opponent of iconoclasts, certainly had miniatures in mind in his famous reference to the painted word and the spoken word standing side by side in ancient manuscripts and both giving the same account of the sacred history. 61 John of Damascus, the other great champion of images, could cite themes that were composed on the order of the great Constantine himself.62 Choricius' description of the mosaics of St. Sergius in Gaza, 63 and Mesantes' discussion of the scenes represented in the Church of the Holy Apostles in Constantinople, 61 both attest monumental cycles of gospel illustrations that go back to the sixth century in the case of the Gaza mosaics, and to the ninth century in the case of those in the Holy Apostles' Church. Two manuscripts of the eleventh and twelfth centuries preserve most completely these ancient gospel sequences: Grec 74 of the Bibliothèque Nationale, with its 372 friezes, and Laur. VI 23 of Florence, with its 288 compositions. M. Millet has urged that these invaluable manuscripts preserve in late and altered form, two distinct and very ancient traditions of gospel illustration. 65 The miniatures of Paris 74 check with Choricius' account of the St. Sergius mosaics in Gaza and relate generally to manuscripts of Syro-Palestinian origin. Here M. Millet finds the Antiochian tradition exemplified. Laur. VI 23, on the other hand, has its affinities with the mosaics of the Holy Apostles in Constantinople and with Alexandrian and Coptic monuments generally. Here, then, the Alexandrian-Constantinopolitan recension is preserved. Comparing the miniatures of Codex 2400 in detail with complete photographs of Paris 74 and the Hautes Etudes selection of the Laurentian miniatures, it becomes evident that the Chicago tableaux belong to the Alexandrian-Constantinopolitan tradition of the latter manuscript, rather than the Antiochian tradition of the former. In subjects treated, in style, and in iconography the parallels to the Chicago miniatures are to be found in the Florentine manuscript. It is a misfortune that the only reproductions of Laur. VI 23 available for students at present are the photographs by M. Millet in the Hautes Etudes collection66 and the sketches by Mme. Millet in Recherches sur /'Iconographie de l'Evangile. Just on the basis of this limited selection alone, however, it is observable that while Paris counterparts for Chicago miniatures are invariably matched by Laurentian parallels, the reverse cannot be affirmed. On the contrary, there are a few Chicago themes that are matched in Laur. VI 23, but have no Paris equivalent. This is surprising in view of the fact that the Laurentian manuscript has fewer friezes than the Paris codex has. In general, also, both the Laurentian and the Chicago miniatures are more directly illustrative of adjacent texts than are the Paris 74 miniatures. 6 1 . Nicephorus, Antirrheticus, I I I , 5; Migue, Patr. Gr., Vol. 100, Col. 382. 62. Joannes Damascenus, De Sanctis et Venerandis Imaginibus, 3 ; Migne, Patr. Gr., Vol. 95, Col. 348. 63. Choricius Gazaeus, Orationes, Declamaliones, Fragmenta, E d . Boissonade (Paris, 1846), p. 94. 64. August Heisenberg, Grabeskirche und Apostel-

kirche,Vo\. 2 Die Apostelkirche zig, 1908).

in

Konstantinopel(Leip-

65. Gabriel Millet, Recherches sur VIconographie de l'Evangile (Paris, 1916), pp. 569-592. 66. Gabriel Millet, La Collection Chrétienne el Byzantine des Hautes Études (Paris, 1903), pp. 39-41, C 3 70-432.

Fi". 29— Qetedjlar, Church : Fresco.

Fig*

30—John's Warning and Subservience; Baptism of Jesus

FIG. 32 — The Last Supper Chicago, Rockefeller McCormick

Baptism ofJesus

[After

Frr,. 31 — The

Japkmiou)

Transfiguration

FIG. 33—John the Baptist Preaching Collection: Byzantine Neto Testament

Repentance

o £

CODEX

2400 A N D I T S

MINIATURES

45

Stylistically there are convincing resemblances between the Chicago and the Florence miniatures with corresponding contrasts in Paris 74. This is most noticeable in the treatment of figures and draperies. F i g u r e proportions are similar in the Chic a g o and Laurentian miniatures, with heads o v e r l a r g e in relation to bodies, and bodies vigorous, stalky, and free moving. In Paris 74 figures are tall, slender, elongated, and normally they stand stiff and erect ( F i g s . 35-36). Draperies in L a u r . V I 23, as in Ms. 2400, are handled freely and loosely. T h e y hang naturally in folds, and whatever decorative quality they possess is unobtrusive. In Paris 74 patterns are repeated and accentuated, borders are straight, and folds rendered in a linear manner, with the lines like the cloisons of an enamel. Of the two styles that of Paris 74 is the more artificial. In the matter of i c o n o g r a p h y there are constant contrasts between L a u r . V I 23 and Paris 74, with the Chicago codex invariably agreeing with the former. Only a few instances—one from each g o s p e l — m u s t suffice. In the Chicago and F l o r e n c e codices J e s u s is seated as he g i v e s instruction« for the feeding of the multitudes in Matthew. In Paris 74 he invariably stands in these scenes (Figs. 35-36). The woman with the hemorrhage kneels in L a u r . V I 23 to touch the garment of J e s u s in illustration of the Marcan s t o r y ; and so she does in the Chicago manuscript. N o t so in P a r i s 74, where she stands conspicuously upright. T h e L u c a n annunciation to M a r y is indefinitely located in L a u r . V I 23 and again in Ms. 2400. In Paris 74 it is the Protevangelium annunciation at the well that is pictured. In the Johannine scene of J e s u s washing the disciples' feet the Laurentian and Chicago codices agree in representing J e s u s as wiping the feet of the disciples. In Paris 74 he is e n g a g e d in the act of washing instead. This collocation of the Chicago cycle with the miniatures of L a u r . V I 23 has undeniable importance for the student of iconography: first, because the Laurentian manuscript, rather than Paris 74, represents the persistence of classical elements in Byzantine g o s p e l illustration; second, because the Laurentian manuscript has notably fewer relatives and descendants than the P a r i s manuscript has. In particular, L a u r . V I 23 does not have close Slavonic parallels such as Mile. D e r Nersessian has discovered and published for Paris 74. 6 ' Indirectly, however, the Chicago miniatures do connect with the Syro-Palestinian tradition represented b y the Paris tetraevangelia. T h r o u g h the medium of Cappadocian influence, that all-important connecting link between the A r a m e a n Orient and Hellenistic Byzantium, the Chicago miniatures reach back toward S y r i a n Antioch. F r o m the time of Asterius of A m a s i a (d. 403), who tells of pictured garments and mural paintings, Cappadocia had been famed for her religious art. In the sixth century some of the greatest E a s t Christian manuscripts that we know were produced in this area. W h e n in the seventh century Syria-Palestine was lost to Christendom b y the A r a b conquest, Cappadocia b e c a m e the great Eastern center for monasticism. In her churches and convents she p r o v i d e d the environment in which a theology and an art tradition derived from the Orient were further developed until the conquest by the

67.

Sirarpie Der Nersessian, Two Slavonic Paral-

lels of the Greek Tetraevangelia: Paris 74, in The Art Bulletin, Vol. IX, No. 3 (March, 1927).

THE ART

46

BULLETIN

Seljuk T u r k s in the eleventh century. Meanwhile, in Constantinople the monastery of the Studium, an unshaken bulwark against iconoclasm, maintained close connections with the monasteries of the provinces, and chiefly with the convents of Cappadocia. Thus through the ramifications of monasticism the Syro-Palestinian tradition and point of view came to have vigorous advocacy in the Byzantine capital. The great monuments of Cappadocian art in the manuscript field are the incomparable purple codices, Rossanensis (S) and Sinopensis (0). More uncouth, but more vigorous and no less sincere or genuinely Cappadocian, are the fragments of an eighth century lectionary in the L e n i n g r a d State Library (Gr. 21). A ninth century Gregory Nazianzenus in the Bibliothèque Nationale (Grec 510) gives evidence of much Anatolian influence in the capital where it was produced. T h e most comprehensive Cappadocian cycles, however, are to be found in the frescoes of the cave churches in Cappadocia itself. 68 It is true that these murals do not date earlier than the ninth century, and most of them are later. But they descend directly from earlier painting under Syrian influence. T h u s they reveal the continuity of the Syro-Hellenistic tradition through this period. Comparing the miniatures of Codex 2 4 0 0 with these m o n u m e n t s of Cappadocian art various evidences of Asiatic influence come to the fore. The lavish pearling of borders was a Cappadocian idiosyncrasy. Bands of pearls are frequently found a m o n g the frescoes in the cave chapels. T h e y stud the furniture in L e n i n g r a d 21 and Paris 510, and they adorn coiffures and headdresses in Sinopensis. In the miniatures of Ms. 2 4 0 0 these pearled borders are found not only on the footstools of the evangelists, b u t also on the footstool and bench of Jesus Himself when Jairus appeals to Him to heal his d a u g h t e r (fol. 42 V, Fig. 40). Painting the g a r m e n t s of Jesus in gold was an item of barbaric splendor that is found in Cappadocian manuscripts. In Rossanensis Jesus' outer g a r m e n t is done in gold while His u n d e r g a r m e n t is blue. In Sinopensis b o t h tunic and mantle are in gold. In the miniatures of Codex 2 4 0 0 the tunic is p u r p l e and the mantle blue usually—but not always. W h e n , in the fifth chapter of Matthew, Jesus addresses the crowd with the encouraging Beatitudes H e is seated on the Mount with a golden mantle w r a p p e d about Him in the manner of the Christ of Rossanensis (fol. 15 v, Fig. 26). In choice of themes represented, also, there is parallelism between the scenes in the Chicago manuscript and the cycles painted in the cave chapels. Cappadocian artists did their creative work in composing gospel scenes. Few illustrations of the Old Testament or of Acts or of h a g i o g r a p h y are included in their repertoire. W h a t is striking is the large increment of apocryphal subjects that are mixed with canonical themes in the gospel cycle. These are derived for the most part from the Protevangelium of James. T h e Virgin Mary is prominent with a cycle of her own, including the presentation of the Virgin, her miraculous feeding by an angel, the reproach of Joseph, the trial by water, a n d the reconciliation of Joseph. Cappadocian artists followed the Protevangelium in picturing the Annunciation, and so represented the Virgin as at her

68.

P. G . d e J e r p h a n i o n , Une novelle

province

de

l'Art Byzantin, Les (Paris, 1925-28).

Églises

rupestres

de

Cappadoce

FEG. 37—Jcstts Jink ring the / liMcrncss

FIG. 3 9 — The Man with a Legion of Demons

FIG. 4 1 — J e s u s Raises J aims' Chicago, Rockefeller

Daughter

FIG, 38—Jesus Teaching from a Boat

Fus. 4 0 — J aims Appeals to Jesus

FIG. 4 2 — T r a n s f i g u r a t i o n Journey

Mc Cor mich Collection : Byzantine

New

Testament

Ftó, 43—Annunciation

to Mary

F i o . 4 4 — D e n u n c i a t i o n of Iferod and Imprisonment oj John

/..

fi

li

/-'/V.A/ .

A >)ffl

Fio, 45-—Nativity

oj John the Baptist

Fie.

46—Nativity with Fluting Shepherd

SP i t ' >V

.>

•5-

Wk F t g . 4 7 — Jesus

Teaching

in

Chicago, Rockefeller

Nazareth McCormick

*M"VfS ' #

rMs ¡ 1

:

f

V

X V

I V -

1, J H

*C R"» ; ' - .• . . .,

i

&I

>

F i g . 4 8 — M i s s i o n of the Twelve Collection:

Byzantine

New

Testament

CODEX

2400 A N D I T S

MINIATURES

4-9

spinning. A l s o in the Nativity scenes appear apocryphal details: the birth in a grotto, the attendant midwives, and the b a b y ' s bath. T h e Cappadocian cycle showed similar apocryphal enrichment in scenes relative to John the Baptist; for in addition to the Nativity of the Baptist there were pictured the pursuit of Elizabeth and the death of Zecharias as details in connection with the massacre of the Bethlehem children. T h e special Baptist interest of Cappadocian artists was further shown in scenes detailing the preaching of the Forerunner and the baptism of J e s u s . All this typically Cappadocian concentration on apocryphal details relative to the Virgin, the Nativity, and John the Baptist is given repeated illustration in Codex 2400. T h e Virgin of the Annunciation in the Chicago codex holds a spindle in her hand and so receives the g o o d news from the angel while at her peculiar work (fol. 57, F i g . 43). Both the Matthean and L u c a n nativities represent the apocryphal birth in a cave rather than the canonical stable scene (fols. 9 and 59, F i g s . 25, 46). A t least one midwife makes her appearance in the Matthean scene. The Chicago codex also contains a Nativity of John the Baptist (fol. 58, F i g . 45) in which the neighbors and relatives of Elizabeth gather to congratulate the mother. T h e S l a u g h t e r of the Innocents in Ms. 2400 is actually the Cappadocian scene of the escape of Elizabeth and her child from H e r o d and his soldiers (fol. gv, F i g . 23). T h e mountain has opened to receive the mother and child, from whom a golden light shines. A s the Protevangelium story requires they are divinely protected in the mountain of G o d (Protev. 23). Pre-Cappadocian examples of this scene are also found in E g y p t i a n frescoes and ivories. E a c h of the gospels in the Chicago manuscript has its Baptist preaching scene in which John preaches repentance or points to J e s u s as the Coming One, as he does in the Cappadocian churches of T o q a l e Kiliss6 and Qeledjlar. 69 F o r the baptismal scene in C o d e x 2400 (fol. i o v , F i g . 30) no less than nine different parallels are to be found among the grotto frescoes of Cappadocia.™ A Qeledjlar mural is typical of the g r o u p (Fig. 29). T h e scene is depicted without perspective or landscape. H e n c e the J o r d a n R i v e r is free to pile up like a tall mound to the v e r y shoulders of J e s u s , as it does also in the Chicago miniature. E l e m e n t s apparently absent from the latter composition but prominent in the Qeledjlar fresco are the dove descending from a segment of heaven above and the trumpeting river g o d J o r d a n in the lower left-hand corner of the watery triangle. In both compositions J e s u s stands v e r y naturally f a c i n g left, not in the frontal pose of the formal middle Byzantine period. H i s hands are not disposed for the concealment of H i s sex, but H i s right hand is raised and extended toward John. F r o m the right two angels approach with bright clothes covering their hands, ready to dry the b o d y of J e s u s . F r o m the left John climbs eagerly upward with bowed back as he does in the g r e a t early representations of the scene on the Monza flasks and the painted panel of the Sancta S a n c t o r u m . This detailed comparison demonstrates a r e m a r k a b l y close resemblance between the Baptism in Ms. 2400 and the Cappadocian treatment of the scene. S u c h similarity between R o c k e f e l l e r McCormick and Cappadocian iconography is not exceptional. Detailed analysis of the Anatolian frescoes has revealed that twenty69. Jerphanion, op. cit., pi. 48, No. 2 ; pi. 65, N o . 1 ; pi. 77, No. i . 70. Jerphanion, op. cit., pl. 35, N o . 2; pl. 36,

No. 3 ; pi. 42, No. 3 ; pi. 48, No. 2 ; pi. 65, No. 2; pi. 77, No. 1 ; pi. 78, No. 1 ; pi. 89, No. 2; pi. 103, No. 3 ; pl. 1 1 9 , N o . 3.

5°

THE ART

BULLETIN

six scenes in the Chicago manuscript are directly paralleled on the walls of the grotto chapels. No less than 139 mural paintings are involved in this comparison. In many instances the resemblance between the miniature and the fresco is nothing short of astonishing. For this reason the Chicago Testament may be ranked with Paris 5 1 0 as one of the products of the imperial scriptorium in Byzantium that shows the persistence of Cappadocian influence in the capital itself. Among the many narrative illustrations of Codex 2400 a scattered cycle of liturgical scenes is clearly distinguishable. These are marked by prominent placement, repetition, formal treatment, non-historical content, and pronounced theological interest. Their presence in a numerous thirteenth century assemblage of text illustrations records one of the most signal and characteristic developments in the whole history of Byzantine art. The earliest East Christian cycles of gospel illustrations appearing in churches and manuscripts were the great narrative cycles, like those in St. Sergius of Gaza or the sixth century antecedents of the Holy Apostles mosaics described by Mesarites. Among the rock churches of Cappadocia this narrative cycle persisted in church decoration until the tenth century. From the time of the Iconoclasts, however, down to the development of the middle Byzantine style in the eleventh century, there was a steady separation of narrative scenes from incidents of the passion and a subordination of the former to the latter. In the end the miracles of Jesus came to be ranked with the legends of saints, and the gospel story was illustrated by a limited selection of scenes especially associated with the great feasts of the church calendar. The explanation of this development in the thought forms of the Byzantines themselves was that the liturgical scenes were absolutely essential to human salvation, whereas the narrative scenes had demonstrative value merely, as proving the divine lordship of the Savior. 71 In this exaltation of liturgical scenes at the expense of the narrative cycle generally, one witnesses the steady adaptation of iconography as well as text to the demands of the Christian cult. The New Testament student, fully familiar with the importance of gospel lectionaries in which the gospel text itself is arranged according to the official calendar of the ecclesiastical year, should not be at all surprised at this parallel ordering of gospel illustrations to celebrate the great church festivals. It is only another illustration of how close were the links uniting liturgy and art in Byzantine history. The process came to a culmination in the eleventh century, sometimes called the " liturgical period." From that time on the schemes of church decoration in Eastern orthodoxy remained fixed, formal, and liturgical. The general arrangement foreshadowed in the Nea or new palace church at Constantinople was perfected and accepted as canonical. It was repeated in the churches of Athos, and Mistra. It is followed even at the present time. Painters' manuals were compiled dictating the precise details both of arrangement and of iconography. These were the precursors of the well-known work of Dionysius of Furna. 72 In manuscript illustrations the 71. 72.

Nicolas Cabasilas, Lit. Expos. Ch. VII. A. N. Didron, Christian Iconography (Tr. by

E. J . Millington, London, 1886), Vol. 2, pp. 189-191, 263-400.

Fio. 4