, science He a al a apes ‘a
SPECULUM
A JOURNAL OF MEDIAEVAL STUDIES_
Volume V, Number 3
July, 1930
Published Quarterly by
THE MEDIAEVAL ACADEMY OF AMERICA CAMBRIDGE, MASSACHUSETTS
EDITORIAL BOARD
Editor-in-Chief JEREMIAH Denis Matuias Forp
Managing Editor Assistant Managing Editor Francis Peapopy Macouwn, Jr SaMUEL Hazzarp Cross
Assistant Managing Editor Publishing Editor CuaRLeEs Hott TAyYLor JOHN Nicuoutas Brown Rupo.iew ALTROCCHI JAMES Hucu Ryan
University of California Catholic University of America
CuHARLES Rurus Morey JoHN StronG Perry Tatiock Princeton University University of California
Dana CARLETON Munro A. A. VASILIEV Princeton University University of Wisconsin
ADVISORY BOARD
Puitie ScHUYLER ALLEN Ricuarp McKEon University of Chicago Columbia University
ALBERT CROLL BauGu CuarRLeES D. MAGINNIs University of Pennsylvania Boston, Massachusetts
CHARLES HENRY BEESON Kempe MALONE University of Chicago Johns Hopkins University
ARTHUR CHARLES LEwits Brown GrEorGE La PIANA Northwestern University Harvard University
CorRNELIA CATLIN COULTER Wituiam ALBERT NITZE Mount Holyoke College University of Chicago
ETIENNE GILSON ArTHUR KINGSLEY PORTER Universities of Paris and Toronto Harvard University
GEORGE LIVINGSTONE HAMILTON Epwarp KENNARD RAND Cornell University Harvard University
CHARLES Homer HaskINs JAMES WESTFALL THOMPSON Harvard University University of Chicago
Wituiam Epwarp Lunt Lynn THORNDIKE Haverford College Columbia University BUSINESS BOARD Amos Puitie McManon JOHN MARSHALL New York University Cambridge, Mass.
GEORGE ARTHUR PLIMPTON New York City
Voi. V, No. 3 — Copyright, 1930, by the Mediaeval Academy of America. — Printep tn U.S. A. Entered as second-class matter, May 8, 1926, at the Post Office at Boston, Mass., under the Act of August 24, 1912.
CONTENTS
Humanistic Studies and Science . . . . . . J. M. Manly
On the Text and Manuscripts of the Speculum Stul- a) ees
The Origins of the Legend of Romeo and Juliet in Se a a eee
The Iconography and the Sequence of the Ambula- tory Capitals of Cluny. ...... . . K. J. Conant
The Textual Criticism of Mediaeval Latin Poets
W. B. Sedgwick
The Gregorian Antiphonary of Silos and the Spanish Melody of the Lamentations ...... . . . C. Rojo
REVIEWS . Terre eee eee ee ee Anniversary Essays in Mediaeval History by Students of Charles
Homer Haskins (J. W. Thompson); I. Bianu and N. Cartojan, Album de Paleografie Roméneascd (Scrierea Chirilicd) (S. H. Cross); F. Holthausen, Beowulf (K. Malone); Ch. Petit Dutaillis and J. Lefebvre, M. I. E. Robertson and R. F. Trehane, transl., Studies and Notes Supplementary to Stubbs’ Constitutional His- tory (W. E. Lunt); E. K. Rand, Studies in the Script of Tours, I; A Survey of the Manuscripts of Tours (H. A. Sanders); E. F. Shannon, Chaucer and the Roman Poets (H. R. Patch); M. R. Toynbee, St Louis of Toulouse and the Process of Canonisation in the Fourteenth Century (J. D. M. Ford); A. de Boiiard, Manuel de Diplomatique Frangaise et Pontificale. 1. (C. H. Haskins).
Announcement of Books Received .
The Editors of Specutum should be addressed in care of the Mediaeval Academy of America, Cambridge, Massachusetts (cable address, ‘Speculum, Cambridgemass.’). The atten- tion of contributors is called to the ‘Notes for Contrib- utors’ printed each year at the end of the January issue.
SPECULUM
A JOURNAL OF MEDIAEVAL STUDIES
HUMANISTIC STUDIES AND SCIENCE!
By JOHN MATTHEWS MANLY
N attempting to study the history of literature I have been in-
creasingly impressed with the futility and even the danger of the boundaries and divisions commonly recognized. Indeed I am per- suaded that such a conception, for example, as the history of Eng- lish literature is not only arbitrary and artificial, but positively detrimental to the pursuit of the object we really have in view. What conception can we form of the intellectual and spiritual life of the English people in the Middle Ages if we confine our study to the writings produced by and for them in the English language? For two hundred years the literature read and enjoyed by the upper classes was that produced at home or on the Continent in French. For an even longer period — in fact, for nearly a thousand years — the language of the most highly educated classes, for all its serious thinking and for no little of its relaxation and diversion, was Latin. And where shall we lay down the boundaries that separate litera- ture, on the one side, from professional writing, and on the other, from journalism? The presence of beauty and skill in construction or expression is not decisive. Many writings of the past occupy an honored place in the history of English literature despite their lack of beauty or power; and the essays of Clutton-Brock were just as fine when they were merely leading articles in The Times as when they were gathered into a book.
But it is not in writing only that the intellectual and spiritual life, the culture, of a people finds expression. The importance of the arts, of law, of government, of institutions, of social ideals and customs, and of industry and commerce needs no emphasis. They
' Presidential address read at the Fifth Annual Meeting of the MepiAEvAL ACADEMY
or America, April 26, 1930. 243
é
244 Humanistic Studies and Science
are essential and important features of human culture and cannot be neglected by the humanist. It is indeed only as the humanist conceives his problems in the light of these facts that his work can have any large and vital human meaning, any claim to equal stand- ing with the natural sciences.
Both the sciences and humanistic studies have their origin in the same impulse — Man’s irrepressible desire to know all that can be known about the world in which he lives and about himself and how he came to be what he is. But despite this unity of origin, it is the belief of many, both humanists and scientists, that there is an im- passable gulf between the sciences and humanistic studies, both in methods and in aims. Some students of literature and literary history look with fierce contempt upon all efforts to introduce scientific methods or aims into these fields of study. Scientific men, on the other hand, have regarded literary scholars, if not with con- tempt, at least with condescension, for two reasons: First, because of a lack of critical methods in ascertaining facts and an arbitrary preference for some other standard than truth in interpreting them. With the growth of historical criticism this cause of reproach has been largely removed and it is recognized, even by scientists, that some humanists apply to the ascertainment of facts a criticism as searching and a logic as inexorable as their own. Moreover, it is recognized also that some of the sciences dealing with inanimate nature must, on the basis of both aims and methods, be classed with the historical sciences; such notably is geology, the aim of which is to ascertain in what order and under what conditions certain se- quences of phenomena occurred in the past and to reconstruct imaginatively the panorama of the life of our planet.
The second cause of reproach still exists, at least in the minds of some critics. Our problems, say our scientific friends, are petty; they have no range, no large significance. Why, they ask us, do you waste so much of your energy upon insignificant authors and periods and phenomena? Why do you not formulate and undertake re- searches of real significance, looking toward the establishment of important and far-reaching general laws, such as have been dis- covered and formulated in the natural sciences?
Humanistic Studies and Science 245
The reply to this is threefold. In the first place, for reasons which need not be discussed at present, humanistic studies as sciences are in their infancy. They are in much the same condition and stage of development to-day that the natural sciences were in during the Middle Ages. They had then existed for a long time. There was a certain inheritance of data and theories wafted down directly from the Greeks and Romans, through tradition and through the meagre stream of classical writers — such a body of learning as we find, for example, in the famous encyclopaedia of Isidore of Seville. This in- heritance was enormously enriched during the twelfth and thirteenth centuries by treatises in all branches of science translated from the Arabic and comprising huge accumulations of Greek learning and the large additions made by Moslem scholars of many lands. These new floods of learning were eagerly received and were tremendously stimulating to some of the most powerful and brilliant minds of that time. There seemed to be no lack of data, there was no lack of in- dustry, and certainly there was a great readiness on the part of scholars to theorize and generalize. But what was the result? Did Albertus Magnus or Peter de Maricourt or Roger Bacon formulate scientific generalizations of importance and far-reaching significance? No; it would be difficult to mention a single general law discovered in the Middle Ages that now forms part of the body of science.
Why was this? The data were abundant, the scientists were able men and ardent scholars. The answer is a thrice-told tale. Al- though the data were abundant, they included as facts innumerable false observations and many mere inventions. More than two hun- dred years passed before it was recognized that science could not ad- vance without casting away all these accumulations and collecting fresh data with critical precautions that would have been incon- ceivable to men of the thirteenth century, and furthermore that generalizations and theories must be derived from the study of the phenomena themselves and not from theology or metaphysics or logic.
In the humanistic fields we have vast accumulations of data — supposed facts — many of which will not bear critical examination and must be replaced by the results of investigations made without
246 Humanistic Studies and Science
bias or prejudice under the strictest observance of critical method. We have many generalizations and theories of widest scope; but un- fortunately most of them were borrowed — not, to be sure, from theology and metaphysics, but from biology and an embryonic and inadequate psychology. They must be cast away and replaced by theories based upon the phenomena themselves.
In the second place the phenomena of our subjects are much more difficult to deal with for scientific purposes than those of the natural sciences. We cannot reproduce them at will under experimental conditions for critical examination. They must be studied when and as they occur. Many of them occurred at times and places so re- mote that only the most meagre and unverifiable information is available concerning precedent and attendant conditions. Even those that are nearer to us in time and space can be known only partially and imperfectly. Moreover, in the whole range of phe- nomena forming the fields of humanistic studies there enter human nature and the human will — elements held by many persons to be incalculable, so irregular in operation as to make absurd any at- tempt to discover in these fields any scientific laws, any invariable sequences of events. These fields, such thinkers tell us, lie outside the reign of law, can never become provinces of the domain of science.
It is not strange, then, that humanistic studies have been slower than the sciences to arrive at any coherent body of laws or even to devise satisfactory methods of investigation. And if it were the task of humanistic science to explain in the individual objects studied — men, books, works of art, and the like — the features which dis- tinguish them from the type or class to which each belongs, the features which constitute their individuality, one might well doubt whether scientific methods and results are possible. But the objects of study of the natural sciences are always and necessarily, not in- dividuals, but types. Formulative science never deals with any- thing else; for the very sufficient reason that only the typical is capable of formulation. In a world that consisted only of disparate individuals science could never formulate, but could only describe, and there would be as many separate descriptions as there were individuals to describe.
a ow ee ee)
Humanistic Studies and Science Q47
This has been the situation with the humanistic sciences in at- tempting to deal with the exceptional — with the elements in the book or the work of art or the institution that set it apart from all the rest. But experience is constantly proving that, even in our respective fields of study, if we attempt to study, not the individual and exceptional, but the typical, regularities can be discovered, ge- neralizations can be made, laws can be formulated.
The ultimate aim of both the natural sciences and the historical is the integration of their fields of study. There is perhaps a dif- ference in the methods of integration proper to each. What form the ultimate integrations of science will take we need not stop to con- sider. For us it seems clear that integration can be accomplished only by the constructive imagination, operating on a secure basis of ascertained fact. Our aim is an intelligible picture of the life of man- kind — its feeble beginnings, the motives and the methods of its creation of instruments, agencies and institutions of a richer, fuller, and nobler existence, the development of its arts and ideals. Part of this task belongs to the anthropologist and archaeologist, part to the orientalist and classicist, part to the modernist, and part to the mediaevalist.
It is hardly too much to say that humanistic studies, with the ideals and purposes which control them to-day, not only are akin to the so-called sciences but are in fact more dependent upon the funda- mental conceptions of science for their very existence than are the natural sciences themselves. When men believed, as the whole Christian world did until recently, that the human race was created less than seven thousand years ago in a single pair of beings whose sin of disobedience resulted in the peopling of the earth by succes- sive generations of men, the historical sciences were inconceivable. The only significant events in the annals of the race were those con- cerned with the punishment, discipline, and training of a chosen people, and with the acceptance of a plan for balancing accounts and at no distant date closing the ledger of human history. The whole past of all the nations but Jews and Christians was a meaningless muddle. The civilizations of Egypt and Tyre and Nineveh and Babylon had flashed into significance for only brief moments of their
248 Humanistic Studies and Science
long night when they served as instruments to aid or discipline the chosen people. China and Japan and India, sitting in darkness, still awaited the moment which would give meaning to their long exist- ence. The sole possible philosophy of history was the story of the fall and redemption of Man. Events which bore no perceptible rela- tion to that might be studied, but only in an idle dilettante spirit, as amusing spectacles or for the utilitarian purpose of extracting from them practical lessons for guidance in similar circumstances.
It is not strange therefore that in the Middle Ages history was never undertaken by men of great philosophical and constructive minds, but was left to the annalists and the propagandists; and that in modern times it attracted only the theologian, the politician, and the polemist. Particular careers, or episodes, or periods could be seen and presented as wholes, subject to some definite principle and disclosing some definite meaning, but the spectacle of the whole movement of the life of the human race as the inclusive subject of humanistic studies was inconceivable until natural law began to re- place magic in explaining sequences of events and the conceptions of theology began to yield to those of biology.
It is in this spirit, I think, that the MepianvaL AcApEMy was born and has worked. Not merely to stimulate many workers to cultivate the mediaeval field, but to aid in developing in each a vision of the whole great process of human life in the Middle Ages is the aim and purpose of the Meprarvat Acapemy. It is not in- tended that the specialist should abandon his specialty and attempt to attack the problem of mediaeval civilization as a whole. To do this would be to revert to the chaotic and unfruitful conditions of mediaeval science. But each specialist will perhaps work with a clearer sense of the dignity and importance of his specialty if he understands and feels that he is helping to write the history of human culture. Seen in this way, for example, the study of the script of Tours becomes, not a mere curious chapter in the development of writing, but a means of tracing, where other evidence fails us, the interrelations of centres of culture in the Middle Ages and some of the channels through which classical literature was distributed. The census of Italian families resident in London in the fourteenth cen-
Humanistic Studies and Science 249
tury becomes interesting not only for the history of international commerce but also for the revision of some of our ideas on inter- national literary relations. The assemblage and study of business terms in use throughout western Europe is a contribution to both philology and the history of commerce and culture.
Millions are being spent on the discovery and reconstruction of the life of man in remote antiquity. We mediaevalists do not be- grudge the orientalists and archaeologists their good fortune. Their task is a great one and the importance of the contributions they have made and are making to the cultural history of mankind cannot be exaggerated. But we do deplore the failure of the intelligent public to recognize in equal measure the claims upon its interest of those great ages which lie so near us and yet are still as deeply buried un- der prejudices and misconceptions bequeathed to us by the Re- formation and the Renaissance as the ancient cities of the Nile and the Euphrates under the drifted sands of Egypt and Mesopotamia. Aside from the enchantment of remote antiquity, it is perhaps the remarkable discoveries of art treasures — buildings, statues, pic- tures, ornaments, fabrics, and documents — that awaken popular interest and secure for oriental research the liberal aid of a host of Maecenases. Unfortunately for the mediaevalist, his treasures have lost their publicity value. Cathedrals not inferior in beauty and splendor to the temples of the East defy the passage of time in many a European city and town, but they have long been known. Carv- ings in ivory, figures in stone and wood, rich designs in gold and silver, enameled and set with precious stones, illuminated manu-
scripts of exquisite beauty, remain in abundance to testify of the skill of the mediaeval artist and the taste of his master, but they are scattered in a hundred museums and treasure-houses; and though their power to charm and refresh us can never fail, they have lost the power to excite us as they would if they had been freshly dug up
yesterday.
This should not be so. Important for the vision of Man’s halting and painful climb upward from the Quaternary mire as are the mysterious civilizations of the Nile valley, the Mesopotamian plain, and the uplands of central Asia, — and no intelligent person can fail
250 Humanistic Studies and Science
to be deeply stirred by the revelations now being made by the ori- entalist and the archaeologist, — the infinitely various and fascinat- ing life of the period we roughly call the Middle Ages must not be neglected. It lies close to us. In it arose many of our most important institutions. Our social life — our customs, our ideals, our supersti- tions and fears and hopes—came to us directly from this period; and no present-day analysis can give a complete account of our civiliza- tion unless it is supplemented by a profound study of the forces and forms of life, good and evil, which we have inherited from it.
To carry out such a program, to construct the large and true picture of the course of human civilization which it is our object to aid in constructing, we need large additions to our funds for special investigations. Such enterprises as the excavations of Cluny are very expensive, but there is no question that Cluny will yield ample returns for all the time and money spent upon it. Equally great and expensive undertakings are the histories of science, and art, and com- merce, and institutions, and of the part played by each in human culture. And as our work grows, it becomes more and more necessary to provide not only for the investigations and the investigators but for a staff of administrative officers free to devote their time and energy and learning to choosing the most fruitful projects and to giving adequate assistance in carrying them on. At present the Executive and Advisory Committees devote to the work of the Academy an amount of time and thought that can be appreciated only by one who knows the number and the magnitude of the pro- jects submitted for their consideration and, if accepted, for advice and active codperation. In particular, the Executive Secretary should be free to devote his whole time to the interests of the Acap- EMY, and the members of the Executive and Advisory Committees should at least have their expenses paid when they attend meetings. That they should give their time and work free is surely all that the Academy can expect. The provision of a permanent endowment is therefore necessary both for investigation and for administration.
UNIVERSITY OF CHICAGO
ON THE TEXT AND MANUSCRIPTS OF THE SPECULUM STULTORUM: II
By JOHN HENRY MOZLEY
N my first article (SpEcULUM, Iv (1929), 430 ff.) I showed that
the readings characteristic of Wright’s MS. B in its divergences from A and C are also found in a number of other MSS, designated by me EFLGHI. Of these EFL contain large lacunae (Wr. pp. 97- 124, and pp. 130 to the end); these I discussed, and gave typical readings for pp. 1-97 in these three MSS and B, and for pp. 1-41 in GHI. I now propose to deal rather more fully with the four MSS BH and GI.
B and H (Arundel 23 and Bodleian 496, F. Madan and H. H. E. Craster’s Summary Catalogue of Western MSS (Oxford, 1922), 0, i, item 2159) show great similarity throughout, both containing the prose Prologue and ending together at Wr. p. 144, 10. We may safely take H’s readings as those of B where B has suffered from loss of leaves (Wr. pp. 41-59). Generally speaking they support EFL, but in the final portion of those MSS (Wr., pp. 124-130) that sup- port I, is varied by readings that agree with the A-group against EFL.
G and I (Digby 27' and Bodleian 780; see Summary Catalogue, loc. cit., item 2583) are also closely akin; they both have the prose prologue, and G ends at Wr. p. 106, 18 ‘dissimulante diu,’ while I has a lacuna at that point until Wr., p. 124, 6 (i e., the end of the first lacuna of EFL), and ends like EFL at p. 130, 16, ‘spreta iacent’; in the middle of the work J ' has lost a number of leaves, from Wr. p. 50, 6 to p. 71, 14.
Both MSS begin by supporting EFL, but after Wr. p. 71 there is a tendency to desert it, though during the final portion of I, which is the same as that of EFL, I is nearer to them than B and H are. The fact that all these four MSS, BHGI, give a general support to the EFL readings, and that between them they contain all the work
! This MS., 1 am informed by Dr Craster, should be dated late 14th century. 251
252 Text and Manuscripts of Speculum Stultorum: I]
down to Wr. p. 144, points to some common ancestor that origi- nated this tradition, and therefore to omission, either purposed or accidental, as the cause of the lacunae in EFL, rather than to those MSS representing the earliest form of the work, and additional pas- sages having been subsequently added. It is not surprising that fifteenth-century MSS should show some influence of the A-tradi- tion. But there is another important consideration, namely that of the lines that are found in the EF L-group but not found in the A- group. It will perhaps be useful to give a table of these:
1. Quam bene reddemus . . . uicem: p. 76, 9, 10. In ELBHGI; not in A- group or F (by A-group I mean the MSS ACDMNJK, for which see below). Probably due to homoioteleuton, though it may be an ad- dition of E.
Committamus . . . tuus: p. 59, 25, 26. In EFLBHG; not in A-group. Though not strictly a homoioteleuton, the double ‘noctis’ may have caused omission by the latter group.
Cellarum . . . parit: p. 84, 5,6. In EFLBHGI; not in A-group. Quamque . . . loqui: ibid. 9, 10. In EFLBH; not in A-group or GI. These seem both to be insertions; there is no special cause of omission, and they can be easily detached from text.
Qua facie ... pedem: p. 85, 19, 20. In EFLBHGI; not in A-group. Pila pusillorum ... ego: ibid. 25, 26. In E...I; not in A-group. The same remarks apply as in Nos 3 and 4.
Lesam .. . recepta: p. 91, 14, 15. In E.... H; not in A-group or GI. Canonicos . . . domus: p. 95,1, 2. In EF BH; not in A-group or LGI. Must have been omitted through homoioteleuton.
Sie igitur ... mihi: p. 95, 11, 12. In EFLBH; not in A-group or GI. One would expect this couplet to follow the next one; in that case
‘ : ° -e . y ) sumat ... sumam may have caused omission, aS In No. 2.
Sed sit ut... reor: p. 100, 1-4. In BHGI; not in A-group. Probably addition; no cause of error.
Murmure .. . simul: p. 112, 19, 20. In BH; not in A-group. Possibly omission through homoioteleuton; text seems doubtful here.
Per me... caro: p. 119, 11, 12. In BH; not in A-group.
Belli . . . prior: ibid., 15-18. In BH; not in A-group. These two seem likely to be later insertions, though the first may be homoioteleuton.
Text and Manuscripts of Speculum Stultorum: IJ 253
. Plurima . . . locutus: p. 133, 4, 5. In BH; not in A-group. Must have been omitted through homoioteleuton.
. Quam uel eos .. . manu: p. 137, 1,2. In BH; notin CDMNJK. No apparent cause of error; perhaps inserted to explain ‘ante’; why ‘eos’?
The lines seem hardly like Nigel’s style.
Of these, Nos 7, 8, 14 are certain omissions, Nos 1, 2, 9, 11 very likely ones. The EFL tradition then goes back to a text earlier than A, for it is a text from which A omitted certain couplets by the com- mon mistake of homoioteleuton. The fact that No. 14 occurs in the second large passage omitted in EFL is an additional reason against explaining those lacunae by the hypothesis of the original form.
Reapincs or BHGI
For Wright, pp. 1-41, see my first article.
For pp. 42-59, where B fails (and I from pp. 50-71):
43 leditur et, ACDJKMN: lesa sed est, EFLGHI
44 illius, ACDJKMN: illata, EFGHI
46 accipiam, ACDJKMN: decipiam; EFGHI
47 cederit, ACJN: sederit, ELGHI
48 after ‘in undis’ EFLGHI have two lines not in the other MSS
51 in urbem, ACDJKMN: ad urbem, EFLGH; uidebunt, A-N: uidebit, E-H
after ‘omne meum’ E-H have 5 lines not in other MSS
53 ait, ACJ: aue, EFH (line omitted in G)
55 multum doluere dolore, ACDJKMN: iam dedi (di) cere dolorem, EFLGH
56 sanum, ACDJKMN: sura, EFLGH
58 profert, ACDJKMN: promit, EFLGH
I have not given all the instances of GHI supporting EFL, but these will suffice to show that support is fairly consistent. The same remark will apply to the next section, pp. 60-97, but here it will be necessary only to refer to the list of readings in my first article, and to point out that the readings in the right-hand column, those of EFLB, are in practically all cases those of H also (the only excep- tions being 74 quos et pudor ipse parentum BHGI with A, ete.
252 Text and Manuscripts of Speculum Stultorum: [7
down to Wr. p. 144, points to some common ancestor that origi- nated this tradition, and therefore to omission, either purposed or accidental, as the cause of the lacunae in EFL, rather than to those MSS representing the earliest form of the work, and additional pas- sages having been subsequently added. It is not surprising that fifteenth-century MSS should show some influence of the A-tradi- tion. But there is another important consideration, namely that of the lines that are found in the EF L-group but not found in the 4- group. It will perhaps be useful to give a table of these:
1. Quam bene reddemus . . . uicem: p. 76, 9, 10. In ELBHGT; not in A- group or F (by A-group I mean the MSS ACDMNJK, for which see below). Probably due to homoioteleuton, though it may be an ad- dition of EF.
Committamus . . . tuus: p. 59, 25, 26. In EFLBHG; not in A-group. Though not strictly a homoioteleuton, the double ‘noctis’ may have caused omission by the latter group.
Cellarum .. . parit: p. 84, 5, 6. In EFLBHGT; not in A-group. Quamque . . . loqui: ibid. 9, 10. In EFLBH; not in A-group or GI. These seem both to be insertions; there is no special cause of omission, and they can be easily detached from text.
Qua facie... pedem: p. 85, 19, 20. In EFLBHGI; not in A-group. Pila pusillorum ... ego: ibid. 25, 26. In E...I7; not in A-group. The same remarks apply as in Nos 3 and 4.
Lesam . . . recepta: p. 91, 14, 15. In E.... H; not in A-group or G/. Canonicos .. . domus: p. 95, 1,2. In EF BH; not in A-group or LGI. Must have been omitted through homoioteleuton.
Sic igitur ... mihi: p. 95, 11, 12. In EFLBH; not in A-group or GI. One would expect this couplet to follow the next one; in that case
‘sumat ... sumam’ may have caused omission, as in No. 2.
Sed sit ut... reor: p. 100, 1-4. In BHGI; not in A-group. Probably addition; no cause of error.
Murmure . . . simul: p. 112, 19, 20. In BH; not in A-group. Possibly omission through homoioteleuton; text seems doubtful here.
Per me... caro: p. 119, 11, 12. In BH; not in A-group.
Belli. . . prior: tbid., 15-18. In BH; not in A-group. These two seem likely to be later insertions, though the first may be homoioteleuton.
Text and Manuscripts of Speculum Stultorum: IT 253
14. Plurima . . . locutus: p. 133, 4, 5. In BH; not in A-group. Must have been omitted through homoioteleuton.
15. Quam uel eos ... manu: p. 137, 1,2. In BH; notin CDMNJK. No apparent cause of error; perhaps inserted to explain ‘ante’; why ‘eos’? The lines seem hardly like Nigel’s style.
Of these, Nos 7, 8, 14 are certain omissions, Nos 1, 2, 9, 11 very likely ones. The EFL tradition then goes back to a text earlier than A, for it is a text from which A omitted certain couplets by the com- mon mistake of homoioteleuton. The fact that No. 14 occurs in the second large passage omitted in EFL is an additional reason against explaining those lacunae by the hypothesis of the original form.
Reapincs or BHGI
For Wright, pp. 1-41, see my first article. For pp. 42-59, where B fails (and J from pp. 50-71): 43 leditur et, ACDJKMN: lesa sed est, EFLGHI 44 illius, ACDJKMN: illata, EFGHI 46 accipiam, ACDJKMN: decipiam; EFGHI 47 cederit, ACJN: sederit, ELGHI after ‘in undis’ EFLGHI have two lines not in the other MSS in urbem, ACDJKMN: ad urbem, EFLGH; widebunt, A-N: uidebit, E-H after ‘omne meum’ E-H have 5 lines not in other MSS ait, ACJ: aue, EFH (line omitted in G@) multum doluere dolore, ACDJKMN: iam dedi (di) cere dolorem, EFLGH 5 sanum, ACDJKMN: sura, EFLGH profert, ACDJKMN: promit, EFLGH
I have not given all the instances of GHI supporting EFL, but these will suffice to show that support is fairly consistent. The same remark will apply to the next section, pp. 60-97, but here it will be necessary only to refer to the list of readings in my first article, and to point out that the readings in the right-hand column, those of EFLB, are in practically all cases those of H also (the only excep- tions being 74 quos el pudor ipse parentum BHGI with A, ete.
254 Text and Manuscripts of Speculum Stultorum: IJ
against EFL, 76 parem GHI with A, etc. against EFLB, 81 nisi quod GI with A etc. cur sic, H: quorsum, EFLB). With regard to GI, however (i.e. G alone from pp. 60-71), the case is different; ¢ supports EFLBH consistently from pp. 60-71, but of the examples given between p. 71 and p. 97 GI only give that support in ‘ Vienna’ for ‘uicina,’ p. 78, and in the addition of the couplet ‘qua facie . .. pedum,’ p. 85; I follows BH in ‘harum sunt . . . partentes,’ p. 94, and G in ‘flagellat’ for ‘flagella,’ p. 97. During the next eight pages GI are if anything closer again to BH; owing to the lacuna they cannot of course be compared with EFL. G then (Wr. p. 106) comes to an end, and I has a lacuna till Wr. p. 124; for these last pages the fol- lowing examples show J allied with EFL against BH:
124 statui reddere, EF LI: statui tradere, H: statuo tradere, AB 125 uidere, ELI: reperere, AFBH 127 hec dicens, ELI: et dicens, AFBH; pondus inerte, ELI: pondere mersa, ABH 128 quam sic ditauit gracia totque dedit, EFLI: cui natura potens tot preciosa dedit, ABH neutra uacat domine seruit utraque sue, EFLI: utraque dat domine debita pensa sue, ABH subsistit, EF LI: subsistens, ABH 129 quicquid, EFLI: quiquam, A: quicquam, BH 130 consuetudo frequens, EFLI: mos malus esse solet, ABH
We have now to consider some MSS of the other class; so far we have had Wright’s A and C, and the MS. which I have called K, but there also belong to this class D/MN.
D (Dublin, Trin. Coll. 440) — for the examination of which I am indebted to the kindness of the College authorities in sending it for me to the British Museum — seems originally to have belonged to a Dublin monk, since on the verso of the 1st folio appears the in- scription in contemporary writing ‘Constat Dompno Petro Le Mona- cho Duutm.’ Its date is about 1400. On the whole it supports A, but agrees with K in many corrections of that MS. It has a singularly large number of clearly impossible readings, from the point of view both of grammar and of metre.
Text and Manuscripts of Speculum Stultorum: JJ 255
J (Oxford, All Souls 37), about the same date as D, is a very close follower of A. This MS., beautifully written in large letters with rubricated headings and initials illuminated in red and blue, is found in the same volume with a text of the Flores Historiarum, which is a copy of that in the Library of Eton College. Eton, however, has no copy of the Speculum, and we cannot say what MS. was copied by J, but it must have been either A or a MS. closely akin.
M and N (Lincoln Cathedral Chapter Library, MSS (105 and 191) come next; here, too, I am indebted to the Librarian, Canon Kynaston, for allowing them to be sent to the British Museum. Of these N is practically a copy of M; it also has the same dislocation that M has (after Wr., p. 126, 14 M leaves a line blank, and then continues p. 135, 26 as far as p. 145, 16; the omitted passage is then inserted, after which follow the last four lines); exactly the same thing occurs in N, though N’s scribe by a marginal note shows that he is aware of the mistake. Further, M has omitted the four lines, Wr. p. 36, 4-11, by homoioteleuton, but inserted them later, and the same thing has happened in N. Again, ten complete lines are omit- ted by both MSS, ‘nec minus . . . ista deus,’ p. 144. N, however, cannot be a direct copy of M, since it has four lines omitted through homoioteleuton by M, Wr. p. 13, 8-11. Also in the first part of the work M has got several rubric headings just one paragraph out of place, but N has not made this mistake (which, however, might be due to correction by NV). The conclusion is that both were trans- scribed from the same original, whose text we have wherever M and N agree.
M (14th century) belonged to the chantry of St Richard de Cante- lupe at Lincoln, N (15th century, Ist half) to the Chapter Library, which explains why N appears in the fifteenth-century catalogue of the Chapter Library while M does not.
While the occasions when C and J differ from A are extremely rare, this is not the case with MNK and D; these four MSS in fact constitute a second tradition of variants from A, though not so marked a one as that of EF and their fifteenth-century supporters. GI are sometimes found in agreement also. Some examples are given here:
256
Text and Manuscripts of Speculum Stultorum: I]
igitur, AC: ergo J:om. DKMN
et ut hoc . . . declarat, ACJ: om. DKMN
iuxta Salomonem qui ait, ACJ: unde Salomonis, DMN: secun- dum illud quod Salomon ait, K
seria, AC: certa, DKMNGIJ
liquida, ACMN: liquide, DJK
donatus, ACJMN: dotatus, DKGI
socialia, ACJ: solacia, DMN: solantia K
nulloque, ACMNJ: nullique, DK
quod ego nunc cunctis, ACMNJ: quod cunctis ego nunc, DK
anona (amoma supr.), A: anona, C: aroma (corr. from annona), J: amoma DKMN
queque tuta minus, ACMN: que caudata minus, DK
obsita, ACJ: consita, DKMN
(rubr.) Confessio Bicornis, AC: Predicacio Bicornis, DK om. J:
Planctus Bicornis (out of place), M: Consolacio Bicornis, N
grauiore, ACMNJ: leuiore, DK
cederit, ACMNJ: cesserit, DK
quid sit ego, AMNJ: quid sim ego, C: sum quid ego (sim, K), DK
uacando, ACMN J: uacabo, DK
Arnoldus, ACMNJ: Arnaldus, DK
Tarabella, ACMNJ: Tarebella, DK
triumpho, ACJ: triumphi, DKMN
after sacrata gerat, DK have 2 lines added.
Discessus Burnelli (rubr.), AJ (om. C): Recessus Burnelli, DKM:
Recessio Burnelli, N.
paribusque, A: paribus sed, K: paribus sed et, D: paribus et, MNJG
nunquam, ACJ: raro, DKMN, sunt et canonici, etc., ACMNJ: DK
have two different couplets for this one.
meo nomen de nomine sumat, ACJ: meum sumat de nomine nomen,
DKMN
reseruant, ACMNJ: reformant, DK
de quibus, ACJ: ipsemet, DKMN
cura, ACJ: crura, DKMN
stertere, ACJ: sternere, DKMN
supplice, ACJ: simplice, DKMN
ac operor mala semper, ACJ: aut operor mala queque, DKMN
Narracio (rubr.), AC: De tribus sororibus fatalibus, DKMN
si Styga iurasset, ACJMN: si de Saturno, DK
pondere mersa, ACMNJ: pondere pressa, DK
fama frequens . . . nouas: this couplet transposed before preceding one in DKMN
Text and Manuscripts of Speculum Stultorum: JJ 257
135 refert, CMNJ: dolet, DK
138 firmata fide, CJM: seruando fidem, DKM
142 uincula cuncta repente, CJ: uincla penultima cuncta, DKMN 144 mazime, CJ: saltem DK
Manuscripts ABROAD
The following MSS exist in countries outside England and Ireland: France. Paris, Bibl. Nat., Latin 16519, fol. 51 ff. Dated 1391. Incomplete, ending ‘fecit et in populo’ ete. (Wr., p. 99, v. 2). Has prose Prologue. Supports EFL group. (S)
Be.cium. Brussels, Bibl. Roy., No. 3156 in vol. v of Joseph Gheyn’s catalogue, ff. 35-94v. Dated 1388, and described on fol. 2 as ‘Liber monasterit Stabulensis.’ Lacks prose Prologue. This MS. and R (see below) were used by Marténe, who in Veterum Scriptorum et Monumentorum Collectio edited a portion of the Speculum (Wr., pp. 81-95) “ex duobus MSS codicibus, uno Stabulensi altero Reginae Sueciae’; and the title of the work in the MS., Gesta Burnelli, agrees with his remark, ‘in veteri codice MS. Stabulensis monasterii opus antiqui poetae invenimus nomine Burnelli.’’ Generally speaking it agrees with the A-group, but more particularly with later MSS such as D and K. The Utrecht and Paris printed editions appear to de- rive from this MS., for among other similarities they share the mis- take of omitting six lines (Wr., p. 39 ‘ne posstt’ to p. 40 ‘ simul’), and the alteration of ‘perurat’ to ‘perurant.’ (0)
Brussels, Bibl. Roy., No. 180, in vol. 1 of Gheyn, ff. 122v.-136v. Dated 1415, and hails from Cologne (‘pertinet fratribus S. Crucis in Colonia’). Lacks prose Prologue, and has the same omission of six lines as O, to which it is closely akin. 460 lines are missing in the earlier part of the work (Wr., p. 12 ‘consuetudo’ to p. 17 ‘tuam,’ and p. 20 ‘ante tamen’ to p. 31 ‘vale’). Belongs to A-group. (P)
Itaty. Rome, Vatican Library, Reg. Lat. 1379, ff. 1-70. The Prefect of the Library informs me that it is early 15th century, and in a hand possibly French, not Italian. It has the words ‘Explicit Speculum Stultorum per manum fratris H. de Palude.’ From the very few variants given by Marténe in the extract quoted by him (see
258 Text and Manuscripts of Speculum Stultorum: [J
above) it may be concluded that this MS. belongs substantially to the same group as 0. Has prose Prologue at end of work. (R) One of the Vienna MSS (V!, below) was surrendered to Italy in 1918, but I have no information as to its present whereabouts. Date, 15th century. (V}) Denmark. Copenhagen, Royal Library, S. 1364, fol. 85-134. Date, 1470. Lacks prose Prologue. Belonged to the Domus Sanctae Mariae in Bordesholm, then to Library at Gottorp. Closely akin to W (q.v.). (Q) Germany. Wolfenbiittel, August-Bibliothek, 616. Helmst., ff. 93-137. Date, 1419. It came to Helmstadt from Liibeck, and had been previously in the possession of Flacius, who had perhaps bought it in England (see M. R. James, Wanderings and Homes of Manuscripts, London, 1919, p. 68). Lacks prose Prologue. It appears to be the MS. referred to by Leyser in Poetae Medii Aen. p. 754: ‘Exstat Sp. St. mstum in bibliothecae Academiae Juliae correctius longe edito exemplari Guelpherbytano [i. e., the Wolfen- biittel edition of 1662], quod ad manus nunc est.’ Belongs on the whole to the A-group, though it would perhaps be safer to call it in- termediate between the two, as it occasionally supports the EFL- readings. @ follows this MS. closely. (W) Breslau, Universitits-Bibliothek, 1V. Q. 126, fol. 153-181. In- complete, ending (Wr. 101) ‘facta nephanda tegunt,’ and has lost a leaf between fol. 163 and 164. Comes from the Corpus Christi com- munity of the Knights of Malta in Breslau. Date, second half of 15th century. Has prose Prologue. Supports the EFL-group. Akin to the first four Vienna MSS (q.v.) and to S. (T) Austria. Vienna, Nationalbibliothek. i. 3154, from Salzburg, 15th century. Now in Italy. (V') ii. 3283. Dated 1418. 87 fol. (V ’) iii. 3467, fol. 1-56. 15th century. (V3) iv. 3487, 44 fol. 14th century. (V 4) vii. 12531, fol. 85-133. 15th century. (V") MSS V!-V4 all agree with 7 in ending at the same line, and in supporting the same tradition. They have the prose Prologue in 4 somewhat shortened form.
Text and Manuscripts of Speculum Stultorum: IJ 259
v. 3529, ff. 1-74v. 15th century. (V 5)
vi. 4459, ff. 2-26v. ca. 1400. (V 8) These two MSS support the A-group. From the couplet at the end of V ®, ‘Spec(u)lum Stultorum reperitur copia quorum/Rome Colonie necnon ubicunque locorum,’ it seems that the MS. may have come from Cologne, and this is supported by the presence of macaronics on the last folio, composed of Latin and Low German mixed. This MS. is also peculiar in writing out the verses as prose, and sometimes substituting a loose prose paraphrase, e. g., for Wright, p. 14, v. 26, we find ‘pro certo enim non est in medico salus quam eger optat’ (= ‘non est in medico quam cupit eger opis’). V ® deals in summary fashion with the prose Prologue, having only the first 12 words.
For the above description of MSS, I have used: (1) information kindly supplied me by the various Librarians; (2) a transcription of the Brussels MS. P made by the scholar Mone, now in the posses- sion of Dr G. G. Coulton, who very kindly allowed me to use it; (3) rotographs of selected portions of QSTW and the Vienna MSS, and of (Q in full.
In none of these MSS is there the long interpolation on the later orders, found in some of the English MSS.
An analysis of all the MSS shows that their allegiance is pretty evenly divided, 14 belonging to the A-group, viz., in England and Ireland ACMNJDK, abroad OPRQWV5V®, and 13 to the other, viz., in England EFLGIBH, abroad STV?%V%V*V". This distinc- tion is admittedly a rough one, some MSS, e. g. G/QW, being of a mixed nature. As for date, we have one of the 13th century, eight of the 14th, four of ca. 1400, and the rest of the 15th century. It is curious that scarcely one of the MSS that support the EF L-readings is complete; H is the sole exception, and we have seen that its support is not consistent throughout. Consequently, although these readings are very often superior to those of the other group, and in some cases date back to a MS. earlier than A, it would be impos- sible to take any of them as a basis for a revision of the text. It seems that A must remain by virtue of its age as the basis, but be corrected where necessary from the other tradition rather than from the later representatives of its own group.
260 Text and Manuscripts of Speculum Stultorum: [/
The dating of the MSS shows that the popularity of the work was greatest between 1350 and 1425, and it may be recalled that Chaucer’s Nun’s Priest’s Tale (CT., B. 4502), which has a reference to ‘Dan Burnel the Asse,’ was written just about the middle of that period.
I have not attempted to construct anything in the way of a stemma of MSS; while undoubtedly providing a pleasing exercise of in- genuity, their practical value appears to be little.
I may be permitted, finally, to refer once more to the question of the lacunae that exist in the three MSS, EFL. These lacunae sug- gested a hypothesis of an earlier and shorter edition of the work, and in my first article (pp. 434 ff.) I pointed out that the text runs on quite naturally from p. 97 to p. 124 (Wr.), and that the inter- vening passage has quite the nature of an interpolation, or rather a later addition; that in the second passage, p. 130 to end, there is nothing connected with the story of the Speculum, that it might not have been in A at all, and that there is no hint of either of these passages in the prose Prologue. Assuming for the moment that the hypothesis is correct, and that Nigel added the passages in question at a later time, we have: First edition of work some time before 1180 (Louis VII of France, referred to Wr. p. 17, died in that year); in 1189 Henry II died, and there was a dispute, in which Nigel took part, between the monks of Canterbury and Archbishop Baldwin (Stubbs, ed., Epistolae Cantuarienses (Rolls Series, No. 38, 1865) pp. 312, 315, 317); Nigel’s treatise Contra Curiales was written after the capture of Richard II in 1192 and before William de Longchamp, Bishop of Ely, had been superseded by Hubert Walter in 1195; William had made himself thoroughly disliked, and the treatise is a bitter attack upon those in authority; about this time then we may imagine the three passages added, viz., the long tirade of Wr. pp. 97-110, the ‘sententiae’ of Crow, Cock, and Sparrow-hawk, and the story of Dryanus. The conversation of the three birds whom Burnellus overhears must in some way repre- sent his criticisms of the laity (‘sed quid de laicis dicam,’ etc.), but it is not clear how, and it cannot be said that there is any indication of date in that passage. The story of Dryanus, however, we find con-
Text and Manuscripts of Speculum Stultorum: JJ 261
nected with Richard I, with whom it was a favorite tale; originally Sanskrit, it seems to have been localised in Italy before 1195: under that year it is told by Matthew Paris as an apologue often repeated by Richard I (see Matthew Paris, Chronica Maiora, Rolls Ser., No. 57, 1874, 1, 413-416, also H. L. D. Ward, Catalogue of Romances . . . in the British Museum, 1, 174; cf. ur (ed. J. A. Herbert), 196). Matthew Paris connects Richard’s liking for the tale with his efforts to get support for the Crusades; it may have begun about then to figure as an exemplum in sermons and other exhortations, so that we are perhaps justified in seeing in its insertion in Nigel’s work a mark of date, namely of the reign of Richard I.
A consideration of the variants in EFL and the rest of that group shows that: (1) in some cases, i.e., those of the lines omitted by A but found in EFL, the tradition goes back to a MS. earlier than A; (2) some of these variants give undoubtedly the true reading, e. g., ‘audito’ p. 5, ‘mutus erat’ p. 11, ‘Stamine’ p. 33, ‘patet’ p. 18; (3) others are either wrong or have the appearance of later improve- ments, e. g. ‘nocutt differre paratis’ (Lucan, Phar., i. 281) in place of ‘noli tardare pedester,’ p. 32, and perhaps ‘orbe quod in toto non futt usque petens’ for ‘nec tamen inuenit que cupiebat bi,’ p. 37, while in other cases there is nothing to choose between the readings; (4) the tradition of variants from A runs right through the other MSS, such as B, that have all or nearly all the poem, suggesting that the original of EFL had it too, and that the lacunae are therefore due to loss of leaves or intentional omission. On the whole this evidence is against the hypothesis, which must therefore be left unproved.
I have also noted the following fragments of MSS:
Munich: Stadtsbibliothek, Cod. Lat. 14529, 15th century, fol. 154: Querela Brunelli ‘Ecce senem cernis fractumque labore decenni’ (p. 97); fol. 160 ‘Brunellus de diversis ordinibus. Ex bibliotheca mon- asterii ord. S. Benedicti ad S. Emmeramum Ratisbonensis.’
Prague: Univ. Library, Cod. III. D. 17, 14th century: ‘fol. prael. non num. Fragmentum Brunelli carminis Satirici Nigelli Wirekert, quod vulgo intitulatur Asinus seu Speculum Stultorum (initio carens).” !
! J. Truhlar, Cat. Codd. Manu. Scriptor. Latinor., 1 (Prague, 1905), 186.
262 Text and Manuscripts of Speculum Stultorum: J]
Cod. X. D. 9, anno 1394: fol. 94b-96b: (Nigelli Wi- rekert) Sp. St. seu Brunellus. (Adest tantum epistola prooemialis ad Wilhelmum et duo versus initiales poematis codice mutilato).!
For completeness’ sake I may mention a late (? 18th cent.) transcription in the Bodleian (Summary Catalogue, No. 3852) made from F, H and J, the two latter MSS being used for the passage Wr. pp. 106-124 and for the Interpolation respectively, and lacking altogether Wr. pp. 97-106 and 124 to the end.
THE INCUNABULA
The Incunabula are not of much importance for the establishment of the text, but it may be of interest to tabulate them (so far at least as I have been able to do so).
1. The Editio Princeps is that of Nicolaus Ketelaer and Gerardus Leempt at Utrecht, about 1474 (= Proctor 8851, Hain 16214); it has 6 lines omit- ted, from p. 39 ‘ne possit’ to p. 40 ‘perurant’ in Wright; fol. 17 should fol- low fol. 6 in this ed.
2. The edition of Gui Marchand at Paris, 1485 or 1490 ( = Proctor 7978, Copinger 6586). Has the same omission as 1.
3. The edition of Conrad Kachelofen at Leipzig, 1487 or 1495 ( = Proc- tor 2904, Copinger 6585); it lacks the prose Prologue, and omiis p. 123 1.5 to p. 130 1.27 in Wright.
4. The edition of Cornelis van Zierikzee at Cologne, February 28, 1499 ( = Proctor 1491, Hain 16218); the date is printed, the pages m iv a and b have been transposed; this edition has woodcuts.
These are all represented in the British Museum. (Hain 16215, 6 and Copinger 6584 I have not been able to examine.) They fall into two groups, 1 and 2 against 3 and 4, 1 and 2 on the whole following the A-group of MSS, 3 and 4 the EF-group.
Since 1 and 2 share the same error with the MS. P (see p. 257), they must either be copied direct from P or from some kindred MS.; the latter seems the more likely, as P is too near to A in its readings; probably some inter- mediate MS. was the source of the Utrecht and the Paris printed editions. Those of Leipzig and Cologne must derive from some MS. akin to the later members of the EF-group, but not now existing, for none of the MSS on the Continent belong to that group, and it is not very likely that either B or H (unless their provenance is foreign) should be the MS.
1 Jbid., 11 (1906), 67.
Text and Manuscripts of Speculum Stultorum: JJ 263
Generally speaking, the incunabula are very free in their treatment of both text and metre (e. g., ‘Cicerone discretior ipso’ for ‘disertior,’ ‘de conche cursu’ for ‘de lucis’ or ‘lincis,’ in Nos 3 and 2 respectively). The Paris edition also alters ‘in titulo caude Francorum rex Lodowycus/non tibi precellit pontificesque sui’ to ‘in caude titulis Anglorum rex Edoardus’ etc.! If this is Edward IV (1461-1483), it is an argument for the intermediate MS. re- ferred to above, since the Paris edition is dated 1490.
Kentons, HasLEMERE, SURREY
THE ORIGINS OF THE LEGEND OF ROMEO AND JULIET IN ITALY
sy OLIN H. MOORE
I
THe MoNTAGUES AND THE CAPULETS IN History
HE powerful families of Montague and Capulet, who figure in Shakespeare’s Romeo and Juliet, never resided in Verona, and in fact never existed at all. Contrary to the general impression, the term Montecchi — the Italian equivalent of ‘Montagues’ — like the term Cappelletti (Cappellini), meaning ‘Capulets,’ was the nickname of a political band or faction, not the name of a private family.’ For the early history of the Montecchi, the most comprehensive and probably the most reliable source is Rolandino, the Paduan notary (died 1276), who chronicles the career of the da Romano family, with especial attention to Ezzelino III. The notary’s work was regarded by contemporaries as so authoritative that it served as the model for several later chronicles, notably the Annales S. Iusti- nae Patavint.?
1 For an opposing view, see Robert Davidsohn, ‘Die Feindschaft der Montecchi und Cappelletti ein Irrtum,’ Deutsche Rundschau, cxvu (1903), 419-428. On p. 420 he speaks of the Montecchi as a Haus, while on p. 426 he says, on the contrary: ‘.. . nun waren die Cap- pelletti keine Familie.’ See also the review of Davidsohn’s article by E. G. Parodi in Bullettino della societa dantesca italiana, x11 (1905), 240-242.
Cf. C. Cipolla, Compendio della storia politica di Verona (Verona, 1900), pp. 130, 131: ‘. .. parte dei Montecchi, famiglia potente, la quale (per quanto sembra) ricevette il suo nome dal grosso villaggio di Montecchio Maggiore, che trovasi tra Verona e Vicenza.’
Cf., however, A. M. Allen, A History of Verona, edited by Edward Armstrong (London, 1910). On p. 45 the writer says: ‘. . . no family called Montecchi ever lived in Verona. It is more probable that the leaders of the party came in the first instance from Montecchio, a hill- fortress in the Vicentine district.’
Miss Allen, while in my judgment more accurate than other writers on this point, does not clearly state that the Montecchi were not a family. She merely says that the alleged family did not reside at Verona. She seems misleading, moreover, in suggesting apparently that the leaders of the Montecchi party resided originally in Montecchio Maggiore. It would seem, as we shall find presently, that during the early part of the thirteenth century the castle was in the hands of the de Pilio (Pillio, or Piio) family, and that it was used by the heterogeneous Montecchi faction merely as their first recruiting station.
2 Rolandini Patauini Chronica, edited by Philip Jaffé, in Monumenta Germaniae His- torica, Scriptores, x1x, 32-147. For the indebtedness to Rolandino of the author of the
264
Origins of Legend of Romeo and Juliet in Italy 265
According to Rolandino, the term Montecchi arose in the following manner: Azzo marquis of Este became podesta of Verona in 1207.! The Ghibelline leader Ezzelino da Romano was violently opposed to such political preferment for his personal enemy. He therefore called a gathering of his followers — from Verona, Vicenza and elsewhere — in the castle of Montecchio. This circumstance served as a sort of christening for Ezzelino’s followers, who were thenceforth known as Montecchi (Monticli or Monticult).?
In another passage Rolandino repeats his assertion that the name Montecchi belonged properly to a political band. He states that Azzo VI, Marquis of Este, had forced an alliance with Bonifacio, Count of S. Bonifacio, and that his faction came to be known as the party of the Marquis. Members of the rival party of Salinguerra Torelli and Ezzelino II da Romano, he says, were known as ‘Mon- tecli.’ *
The castle of Montecchio Maggiore, the ruins of which are still visible, was situated in the neighborhood of Montebello Vicentino, in the foothills of Vicenza, and is the only authentic landmark for the story of the Montagues and the Capulets. As was indicated by the Latin form of its name — Monticulus, ‘little mountain’ — it was
built upon a small eminence. It matters little that eventually the name of the castle seems to have been adopted by members of the family resident there — ‘Marchixius and Confortus de Monticulo’ ‘
Annales S. Iustinae Patavini, see ibid., p. 148, and especially n. 3. The Annales S. Iustinae Patavini were published under the title of Monachi Patauini Chronicon by L. A. Muratori, Rerum Italicarum Scriptores (Milan, 1727), vii, 665-734.
1 Parisius de Cereta gives the date as 1206 (Monumenta cit., Annales V eronenses, xx, 6).
2 ‘|. congregatis amicis suis de Verona Vicencia et undecumque potuit aliunde in cas- trum Monticli Vicentini districtus — unde postea dicti sunt Monticli . . .” — Rolandino, op. cit., p. 43.
Rolandino’s statement concerning the origin of the Montecchi was copied, in somewhat garbled form, by Lorenzo de’ Monaci: ‘. . . Ezerinus . . . congregatis omnibus amicis suis de Verona, Vicentia, et Monticulo (unde postea Monticuli dicti sunt)...’ — Laurentii de Monacis, Ezerinus III, Bk. xiii, in L. A. Muratori, op. cit., viii, 138.
’ ‘Et alii fovebant partem comitis de Sancto Bonifacio, que erat pars marchionis; alii partem Sallimwerre, que erat et Eccelini, et hi dicebantur Montecli . . .”— Rolandino, op. cit., p. 50.
‘ ‘Item eo anno (1285) dictus potestas Parme armata manu cum 2000 peditibus, banderiis levatis, tubis et campanis pulsatis, ivit Monticlum ad domum Marchixii et Conforti de Monti- culo.” — Annales Parmenses, edited by Philip Jaffé, in Monumenta cit., xvi, 698.
In 1247 troops from Parma retreated to this castle, and thence continued their march to
266 Origins of Legend of Romeo and Juliet in Italy
— just as the family of Ezzelino da Romano itself derived its name from the ancestral castle of Romano.’ The essential fact is that we find no authoritative mention in the first half of the thirteenth cen- tury of a da Montecchio family. For instance, in the year 1239, ac- cording to Rolandino, the castle of Montecchio belonged to Ugucio de Pilio (Piio, or Pillio), and was captured by Emperor Frederick IT.? In 1242, it appears, Ugucio surrendered his castle to Ezzelino da Romano.’
It is perhaps significant also that the Monticuli are at first in- variably referred to in the plural, as if indicating their conglomerate nature, recruited as they were from all parts of northeastern Italy, or undecumque, as Rolandino significantly observes.
Not only was the term Montecchi reserved at the outset for a political band in Verona, but the term Quattuorviginti (Italian Quat- troventi), applied to the chief allies of the Montecchi, was apparently also the nickname of a faction. Quattuorvigintt seems to have meant originally the eighty followers of the Count of S. Bonifacio who were bribed by Ezzelino da Romano and Salinguerra Torelli to become renegades.‘ While this theory rests upon no more secure basis than a marginal note in one of the manuscripts of the Annales S. Iustinae Patavini, it affords nevertheless the most plausible explanation of- fered for the source of the name.*
On the other hand the Counts of S. Bonifacio — the chief anta- gonists of the Montecchi and the Quattroventi — represented a real family, descended from Milo, Count of Verona, who ruled from 930 to 950, and from his brother Egelric.6 The story goes that a certain Tiresio killed Count Sauro, of this family, on the hill of S. Bonifacio. the city of Parma (ibid., p. 672). In 1296 the castle was stormed by Azzo Marquis of Este, who
was nevertheless obliged to withdraw, because of the loss of seven of his best soldiers (ibid., p. 719).
1 Gerardi Maurisii Historia, in L. A. Muratori, op. cit., vi, 14B. The da Romano family formerly went by the name of da Onara. Cf. also W. F. Butler, The Lombard Communes (New York, 1906), p. 207.
2 ‘Et (imperator) stans ibidem Vicencie, fecit sibi Monteclum maiorem dari, castrum scilicet Ugucionis de Piio Vicentino districtus, et positis ibi custodibus Sarracenis, fecit ipsum ulterius per imperium custodire.’ — Rolandino, op. cit., p. 72.
3 Parisius de Cereta, op. cit., p. 12. 4 A. M. Allen, op. cit., p. 51, and n. 2.
5 Annales S. Iustinae Patavini cit., p. 152, and n. d.
6 A. M. Allen, op. cit., pp. 11, 45.
Origins of Legend of Romeo and Juliet in Italy 267
Tiresio left for the Orient after the homicide, but his brothers made peace with the heirs of Sauro by ceding to them the hill of S. Boni- facio, near Verona.' Thereafter the Counts of Verona took the name of S. Bonifacio.
For the first third of the thirteenth century, the chronicles are replete with accounts of the battles of the quarrelsome Montecchi, whose fluctuating fortunes it will be impossible to follow in detail here. After 1236, mention of the Montecchi becomes much rarer, but apparently their power, as well as that of their allies the Quattroventi, still continues in 1245, when a distribution of enemy booty was made in their favor.”
Under date of 1252 occurs what is apparently the first attempt to use the term Montecchi as a family name. Parisius de Cereta, who seems to have had a fondness for adding titles to proper names, makes a bare reference to a certain Carnarolus de Monticultis as one of the numerous victims put to death by Ezzelino da Romano in that year.? Rolandino, however, who gives a detailed account of this af-
1 C. Cipolla, Compendio cit., p. 131.
The anonymous author of Ricciardi Comitis Sancti Bonifacit Vita (Muratori, op. cit., vit, 122) makes the murder occur in 1184, the victim being ‘Alexandrum Comitem San-Boni- facium’ and the murderer a member of the Montecchi ‘family,’ Ceresius ‘Monticulus.’ This utterly unreliable ‘chronicler,’ who is undecided about the Trojan origin of the S. Bonifacio family, makes Bonifacio the son of Sauro, and the nephew of the murdered ‘ Alexander’ (ibid). According to him, Ludovico was the son of Bonifacio, and Rizardo the son of Ludovico.
‘...et clm Guelphorum pars praevaluisset, virtute potissimim Ricciardi, Ludovici S. Bonifacii Filii, tune pené imberbis, impigri, ingentisque spiritus adolescentis, Monticulis Urbe pulsis, Azo Estensis Marchio, auctore Ludovico Ricciardi patre, Praetor Veronae crea- tur...’ (ibid.). Cf. col. 123: ‘. . . Ludovicus Sanbonifacius cum filio Ricciardo . . .’
As a matter of fact, Ludovico was the son, not the father, of Rizardo, whose biography this anonymous ‘chronicler’ essays to write. ‘... (Ricardus) ... Reliquit autem unicum filium parvulum nomine Ludovicum . . .. — Annales S. Iustinae Patavini, p. 162. Cf. A. M. Allen, op. cit., p. 80. Rolandino refers to this son as ‘Leoisius’ (op. cit., p. 81).
It might be noted also that Parisius de Cereta makes Bonifacio the son of Sancius (= Sauro?) of S. Bonifacio (op. cit., p. 6).
The date of the death of Sauro is given in the Annales Sanctae Trinitatis as 1189 (Monumenta cit., x1x, 5).
2 *1245 ... Veggasi doc. 26 apr. 1245 S. Mich. in Camp. 387, dove sotto la podesteria di Guiberto de Vivario si parla de distribuire “bona et possessiones Jnimicorum inter amicos partis monticulorum et quattuorviginti.”” —C. Cipolla, Antiche Cronache Veronesi (Ve- nice, 1890), p. 292, note.
* Parisius de Cereta, op. cit., p. 14. Parisius de Cereta not only gives a title to all the other persons mentioned in this passage, but bestows the title of comes Sancti Bonifacii upon Marcus Regulus, who died in 1142, and is referred to in the Annales Breves simply as comes Malregula-
268 Origins of Legend of Romeo and Juliet in Italy
fair, mentions no connection between Carnarolus and the Montecchi, According to his version, ‘Dompnus’ Carnarolus, as well as his brother, Friar Felisius, of Padua, were victims of Ezzelino III da Romano. He calls Carnarolus a prominent citizen — ‘potentem virum et divitem’ — of Verona. His brother, on the contrary, seems to have been a humble priest, and a teacher of the Germans in Padua: ‘Praeceptor sive magister Alemanorum in Padua.’! Carnarolus was imprisoned, compelled to do menial service in the house of the tyrant, and then publicly flogged in the streets of Verona.*
While it does not seem, consequently, that the title de Monticulis given by Parisius de Cereta was well established, there appears nevertheless to have been a gradually increasing tendency to use this sobriquet as part of a family name. For instance, on September 23, 1279, one of the signers of the peace between Brescia, on the one hand, and the allied citizens of Verona and Mantua on the other, was the Mantuan Astolfinus qui dicitur Monteclus.* Again, in 1324, according to an unedited chronicle, Crescimbene de’ Monticoli with his two sons was expelled from Verona by Can Grande della Scala, and finally took refuge in Udine.‘ This gradual application of a party nickname to private families may be compared with such American expressions as ‘Gypsy’ Smith, ‘Silver Dick’ Bland, or ‘Texas’ Jones, indicating the tribal, political, or geographical connections of the individual.®
tus (Monumenta cit., xix, 2). The best evidence would indicate that the title of comes Sancti Bonifacii was not adopted by the family until the latter part of the twelfth century. See C. Cipolla, Compendio cit., p. 181. 1 Rolandino, op. cit., p. 98.
2 ‘Supra dictus siquidem dompnus Carnarolus in platea Verone est dilaceratus per frusta ...’ (ibid., p. 99).
In the Annales S. Iustinae Patavini, Carnarolus is spoken of simply as a ‘ miles Veronensis,’ and a victim of Ezzelino da Romano (Annales S. Iustinae cit., p. 162).
3 C. Cipolla, Documenti per la storia delle relazioni diplomatiche fra Verona e Mantova nel secolo xiii (Milan, 1901), p. 178.
4 “Non senza amaritudine singhiozzando rammento lo sventurato giorno, nel quale il magnifico M. Crescimbene de’ Monticoli con due figli di tenera eta, e con il signor Federico Della Scala (di cui Crescimbene favori le trame contro Cangrande), fu nel 1324 crudelmente scacciato da Verona. Al quale dopo lunga peregrinazione, da Carlo IV imperatore fu concesso di abitare in Udine (1348), e fu provveduto di onorevole mantenimento.’ — Cronaca Ms. Udinese, cited by Cino Chiarini, Romeo e Giulietta (Florence, 1906), p. 42, n. 2.
5 In the Appendix to the 14th century Antonii Godi Nobilis Vicentini Chronica, in L. A. Muratori (op. cit., v1, 91, 92), is a list of Vicenza families which have been lost sight of:
Origins of Legend of Romeo and Juliet in Italy 269
If the historical development of the term Montecchi has been little understood in the past, we have been more fortunate with re- gard to the term Cappelletti. Thanks to the valuable researches of R. Davidsohn, the origins of the Cappelletti are fairly familiar to specialists in Italian mediaeval history, although still ignored, unfor- tunately, by certain authorities on the literary history of the period. For the benefit of the latter, it will be recalled here that the Guelf Cappelletti, far from being a family living in Verona, were a faction associated with the political affairs of Cremona. Their Ghibelline opponents were known as Barbarasi,' according to reliable chron- iclers, or Troncaciuffi, according to the commentator Peter Ali- ghieri.2 Obviously the nicknames Barbarasi and Troncaciuffi referred to the practice of shaving their beards, as a mark of distinction.* In fact, they have been compared appropriately to ‘the Round- heads and the Croppies of English and Irish history.’ The Cappel- lett, or Cappellini, for their part, evidently wore small hats or caps as insignia. Later, as was pointed out by an able Italian critic three quarters of a century ago, history repeated itself, and the name was adopted by a troop of Venetian light horsemen — a fact which may have had its weight with Luigi da Porto.‘
The fortunes of the Cappelletti may be traced briefly, if only to establish further the parallel with the fate of the Montecchi. In 1249, taking advantage of the declining authority of Emperor Frederick ‘Hae sunt familiae, quae in Civitate nostra Nobiles erant, et ita extinctae ut de eis vir maneat memoria.’ Among these ‘families’ were mentioned the ‘Comites Monticuli Praecalcini,’ who
were given as ancestors not only for the Montecchi of Verona, but also for their bitter rivals, the Counts of S. Bonifacio.’
‘Comites Monticuli Praecalcini, ex quibus nati sunt Comites de Sancto Bonifacio, et Domus Monticulorum qui potens fuit Veronae.’
1‘. ...habendo Amadinus magnam partem in civitate que Capelleti appellabatur; alii intrinseci Barbarasi dicebantur . . .” — Annales Placentini Gibellini, in Monumenta cit., xvi, 498.
* Petri Allegherit Super Dantis Ipsius Genitoris Comoediam Commentarium, edited by Vincenzo Nannucci (Florence, 1845), pp. 332-333.
? See R. Davidsohn, loc. cit.; W. F. Butler, op. cit., p. $49.
* Cf. Giuseppe Todeschini, ‘Lettera a Bartolommeo Bressan,’ in Lettere storiche di Luigi da Porto, edited by Bartolommeo Bressan (Florence, 1857), p. 428.
For Cappellini, the alternate form for Cappelletti, see Cronaca di fra Salimbene Parmigiano del? ordine dei minori, translated by Carlo Cantarelli (Parma, 1882), 1, 264: ‘In Cremona, que’ che parteggiavano per la Chiesa si chiamavano Cappellini, o Cappelletti; que’ che tenevano per ’ Impero, si nominavano Barbarasi.’
270 Origins of Legend of Romeo and Juliet in Italy
II, the Cappelletti attempted an uprising against the Barbarasi, without success. The insurgents were banished from Cremona, to- gether with their leader Amadinus de Amatis.'
The following year, a band of some 2000 Barbarasi attacked the Cappelletti in their place of refuge. After a long siege, Amadinus de Amatis was captured. By a stroke of good fortune, the prisoners were all permitted to escape by their custodian Marquis Lancia, podesta of Lodi. The cause for this leniency remains a disputed point — whether bribery, or sympathy with the Cappelletti faction, or jealousy of the podesta of Cremona.’
In 1259, we find the combined authorities of Verona, Mantua, Ferrara and Padua making an alliance with the podesta of Cremona, by the terms of which they agree to consider the Barbarasi party as being the commune of Cremona, and promise to give neither aid nor counsel to the Cappelletti banished from that city.*
In 1267, it would appear, the Cappelletti had recovered some of their lost power. Through the intercession of a legate of Pope Clement IV, they were allowed to return to Cremona,‘ where they seem to have caused few disturbances worth recording. The last reference to them by a contemporary occurs in Dante’s Purgatorio.
As far as authentic records go, the Ghibelline Montecchi of Verona never seem to have come into conflict with the Guelf Cappelletti of Cremona. The history of the twelfth century, however, furnishes a curious parallel for the formation of their names. In that century, the mercenary troops commonly known as Brabancons derived their names from the duchy of Brabant, where they were first recruited, much as the Montecchi were named for the castle of Montecchio Maggiore, first recruiting station of Ezzelino II da Romano. The antagonists of the Brabancons were called Caputzi, from the linen hoods which they wore, just as the Cappelletti were named also for their characteristic headgear. Apparently, no claim has ever been advanced that the Caputii were a private family.
1 Annales Placentini Gibellini, loc. cit. 2 Tbid., p. 499. 3 C. Cipolla, Documenti cit., pp. 71, 77. * Annales Parmenses Maiores, p. 680.
Origins of Legend of Romeo and Jultet in Italy 271
II
EARLY DEVELOPMENT OF THE LEGEND OF THE MonTAGUES AND THE CAPULETS
If it be admitted that the Montecchi and the Cappelletti existed as unrelated political factions, some of the currently accepted views regarding the origins of the legend of Romeo and Juliet will have to be revised. It will be impossible to accept, for instance, the view of Cino Chiarini, who says, following Todeschini: ‘I Montecchi e i Cappelletti nominati da Dante rappresentano due diverse frazioni del partito ghibellino.’! The Cappelletti, as we have seen, were fol- lowers of the Pope, not of the Emperor. Nor will it be possible to accept the statement of G. Brognoligo, whose views still find the widest acceptance, that the Montecchi and the Cappelletti were a ‘coppia di famiglie.’ *
In fact, when we come to trace the literary tradition of the Montec- chi and the Cappelletti, we shall be confronted by a new problem, similar to that presented in the historical evolution of the Montecchi. It will be our task to determine at what stage of the development of the legend the idea arose that the Montecchi and the Cappelletti were families. A second problem will be: When did the notion of hostility between the Montecchi and the Cappelletti first manifest itself?
To answer these questions, for which no satisfactory solution has yet been offered by literary critics, we shall be obliged to begin at the source, and trace the early development of the legend step by step.
The earliest apparent association of the terms Montecchi and Cap- pelletti occurs in the sixth canto of the Purgatorio. Here Dante, in- spired by the stately presence of the troubadour Sordello, pro- nounced a bitter apostrophe to Italy, in which he invoked the aid of Albert of Hapsburg. He earnestly beseeched this emperor, who was so indifferent to Italian affairs that he had neglected to be crowned
1 W. Shakespeare: Romeo e Giulietta, edited by Cino Chiarini, (Florence, 1908 or later),
. Xi. , * Gioachino Brognoligo, ‘Montecchi e Cappelletti nella Divina Commedia’ (Bologna, 1893), Estratto dal Propugnatore, Nuova Serie, v1, Part 1, Fasc. 31-32, p. 8.
272 Origins of Legend of Romeo and Juliet in Italy
in Italy, to come and behold the civil strife which was demoralizing
the country: Vieni a veder Montecchi e Cappelletti, Monaldi e Filippeschi, uom senza cura; Color gia tristi, e questi con sospetti! !
With his characteristic symmetry, Dante balanced the names of the Montecchi and Cappelletti, factions which had destroyed them- selves in past strife (color gid tristi), with the contemporary and bet- ter known parties — the Monaldi and Filippeschi of Orvieto — who were in a dubious situation because of their dissensions (e questi con sospetti).2 The language of the poet was somewhat cryptic, as was customary when he assumed the réle of high priest. Furthermore, the first two factions mentioned had completely passed from popular recollection. As a consequence, a series of misinterpretations arose, which became crystallized into one of the most famous legends of literature.
Let us now examine, in chronological order, some of the early comments upon this passage.
(a) Jacopo Deis Lana
1323-28. — According to Jacopo della Lana, who obviously under- takes little more than a paraphrase of the text of Dante at this point, the Montecchi and Cappelletti were a party in Cremona, while the Monaldi and Filippeschi were a party in Ancona.* Apparently the first-named factions were allies, in his opinion, although they might also have been taken to be enemies.‘
(b) Ortimo CoMMENTO
1334. — The Ottimo Commento here follows literally Jacopo della Lana.°
1 Orvieto was in the province of Ancona (ibid., p. 14, and n. 8).
2 ‘Qui per digressione noma parte di Cremona per principio di Lombardia, e parte d’An- cona per principio della Marca Anconitana.’ — Comedia di Dante degli Allagherii col Com- mento di Jacopo della Lana Bolognese, edited by Luciano Scarabelli, in the Collezione di Opere Inedite o Rare dei Primi Tre Secoli della Lingua (Bologna, 1866), 11, 72. Scarabelli changes the reading Cremona, given by all the codices, to Verona.
3 Purgatorio, vi, 106-108. * See G. Brognoligo, op. cit., p. 14.
5 L’Ottimo Commento della Divina Commedia (Pisa, 1828), 11. 83-84.
Origins of Legend of Romeo and Juliet in Italy 273
(c) Prrer ALIGHIERI
1340-41. — The first commentator furnishing accurate historical information regarding the political situation in Verona and Cremona during the thirteenth century was Peter Alighieri. According to him, there were two parties in Verona, that of the Montecchi, and that of the ‘Counts.’ In Cremona, the Cappelletti were opposed by the Troncaciuffi. In Orvieto, the rivals of the Monaldi faction were the Filippeschi.'
By the party of the ‘Counts,’ Peter Alighieri doubtless means the faction of the Count of S. Bonifacio, which was also called the party of the marquis, because one of its leaders was the Marquis of Este. The name Troncaciuffi, as has been observed, was an alter- native form for Barbarasz.
(d) Copice CAssINESE
Middle of 14th century. — The author of the Codice Cassinese follows closely the commentary of Peter Alighieri.’
(e) GIovANNI VILLANI
i348. — The chronicler Giovanni Villani does not mention the Montecchi and Cappelletti, but describes the Guelf Monaldeschi and the Ghibelline Filippeschi as families which were active rivals in the city of Orvieto during the early part of the fourteenth century. In 1312 the Monaldeschi succeeded in expelling the Filippeschi from Orvieto, almost under the nose of the Emperor himself. In 1337, after a period of tyrannical rule, the Monaldeschi were themselves driven out by the indignant citizens of Orvieto.’
1 ‘De quo secutum est quod in Verona est facta pars Montecchia et pars Comitum; in Cremona Cappelletti et Troncaciuffi; in Urbeveteri pars Monaldeschia et Philippesca; et sic de aliis.’ — Peter Alighieri, loc. cit. See G. Brognoligo, op. cit., p. 10.
* Il Codice Cassinese della Divina Commedia, edited by Monaci Benedettini, Tipografia di Monte Cassino (1865), p. 220.
For paleographical reasons the editors date the commentary about the middle of the fourteenth century (ibid., pp. xviii and x1).
* Cronica di Giovanni Villani (Florence 1823), 1v, Bk.1x 8, chap. xl, 34, 35; and vi, Bk. xi, chap. Ixxv, 153.
274 Origins of Legend of Romeo and Juliet in Italy
(f) Benvenuto pa Imoia
1379. — Benvenuto da Imola, for the passage in question, utilizes several sources. Like Giovanni Villani, he asserts that the Monaldj and Filippeschi were noble houses of Orvieto, but goes a step further, originating the statement that the Montecchi and Cappelletti were families also. This natural error may doubtless be ascribed to the influence of Villani.
Benvenuto gives an account of the alliance of the Montecchi with Ezzelino da Romano, and of their struggle against Azzo Marquis of Este, who succeeded in returning to Verona, owing to the support of the Count of S. Bonifacio. Except for one important detail, he fol- lows here the Monachi Patavini Chronicon, or Annales S. Tustinae Patavini.' He differs from the chroniclers, however, by representing the Cappelletti as allies of the Montecchi, a notion which he may have derived from Jacopo della Lana.’
(g) Francesco pa Buti
1380. — Probably influenced by Jacopo della Lana, Francesco da Buti places both the Montecchi and the Cappelletti in Cremona. He
is the first writer to mention hostility between the two parties, hay- ing perhaps some inkling of their difference in political faith.’
(hk) Comento ms. Tratto pA Varys Curosatori (Barberiana de Roma)
14th century? — The above-mentioned commentary, which is cited by Alessandro Torri, represents the Montecchi and Cappelletti
1 Op. cit., p. 149. See John M. Gitterman, Ezzelin von Romano, I. Teil: Die Griindung der Signorie (1194-1244) (Stuttgart, 1890), p. 94.
2 ‘Dicit ergo: Vieni a veder Montecchi e Cappelletti. Istae fuerunt duae clarae familise Veronae, maxime Monticuli, quae habuerunt diu bellum cum alia nobilissima familia, scilicet, cum comitibus de Sancto Bonifacio. Nam Monticuli comites cum favore Eccerini de Romano eiecerunt Azonem II, marchionem estensem, rectorem illius civitatis; sed ipse in manu forti cum comite Huberto Sancti Bonifacii, Monticulis acie debellatis, reintravit Veronam, uwbi finem vitae feliciter terminavit. Et dicit: Monaldi e Filippeschi; istae fuerunt duae nobiles domus de urbe veteri...’ — Benvenuto de Rambaldis de Imola, Comentum super Danlis Aldigherij Comoediam, Sumptibus Guilielmi Warren Vernon, curante Jacobo Philippo Lacaita (Florence, 1887), vol. 11.
3 Commento di Francesco da Buti sopra la Divina Commedia di Dante Allighieri, edited by
Crescentino Giannini (Pisa, 1860), 1, 138.
Origins of Legend of Romeo and Juliet in Italy 275
as hostile factions, after the manner of Francesco da Buti. By a curious inversion, however, the author makes the Montecchi Guelfs, and the Cappelletti Ghibellines.'
(t) ANOoNIMO FIORENTINO
Late 14th or Early 15th Century.— The Anonimo Fiorentino, abridging the commentary of Benvenuto da Imola, asserts that the Montecchi and Cappelletti were families allied against the Marquis of Este in Verona.’
(7) Epizione NiposBeaTINA
1478. — The author of the Edizione Nidobeatina follows closely Peter Alighieri. His geographical explanations are taken apparently from Jacopo della Lana.°
(k) Curistororo LANDINO
1481. — For the passage under consideration, Christoforo Landino abridges the commentary of Benvenuto da Imola.‘
(1) ALESSANDRO VELLUTELLO
1544. — Alessandro Vellutello, as usual, repeats the statements of Christoforo Landino.°
(m) BERNARDINO DANIELLO
1568. — Following Benvenuto da Imola, Bernardino Daniello represents the Montecchi and Cappelletti as being allied families. He is the first commentator to speak of the Cappelletti as Ghibel- lines.®
1 Cited by Alessandro Torri in Giulietta e Romeo (Pisa, 1831). p. xviii.
? Commento alla Divina Commedia d’ Anonimo Fiorentino del Secolo XIV, edited by Pietro Fanfani (Bologna, 1866-1874), 11, 107.
4 Alighieri Commedia con Commento (Milan, 1478), Fo. 97r.
* Dante con l'Esposizione di Christoforo Landino e d’ Alessandro Vellutello, edited by Francesco Sansovino Fiorentino (Venice, 1596), p. 186.
5 Ibid., p. 187.
* Dante con l’ Esposizione di M. Bernardino Daniello da Lucca, (Venice, 1568), p. 274.
276 Origins of Legend of Romeo and Juliet in Italy
We have now reviewed the early stages of the development of our legend, and the results may be summarized as follows:
The traditional conception of the Montecchi and the Cappelletti really begins with an obscure passage in Dante’s Purgatorio, where, in one verse, the names of the factions are found in juxtaposition. In the verse immediately following, the poet.places the names of rival factions in Orvieto, the Monaldi and the Filippeschi.
Three interpretations for the passage presented themselves:
(a) The Montecchi and Cappelletti were allies, and probably Ghibellines, because they were described as gid tristi, and hence were proper subjects for the Emperor’s solicitude.
(b) The Montecchi of Verona and the Cappelletti of Cremona were early thirteenth-century examples of civil dissension, while the Monaldi and Filippeschi were early fourteenth century examples of the same evil.
(c) The Montecchi and Cappelletti were hostile factions — or perhaps families — just as the Monaldi and Filippeschi were said by Villani to be rival families.
The conception of the Montecchi and Cappelletti as allies, first ap- parently suggested by Jacopo della Lana, was developed by Ben- venuto da Imola, who originated the statement that the names in question belonged to families. Benvenuto also drew from Paduan chronicles details of the conflict between Ezzelino da Romano and the Este family. His ideas were later adopted by Anonimo Fioren- tino, Christoforo Landino, Alessandro Vellutello and Bernardino Daniello.
Peter Alighieri was the first commentator to explain that the Montecchi were a faction of Verona, while the Cappelletti were a party at Cremona. The author of the Codice Cassinese copied him, while the author of the Edizione Nidobeatina amplified his state- ments. Peter Alighieri’s explanation, while historically correct, proved the least popular of all.
The idea of hostility between the Montecchi and Cappelletti, first suggested by Francesco da Buti, is found in the Comento MS. Tratto da Varj Chiosatori. Eventually this innovation was generally
Origins of Legend of Romeo and Juliet in Italy 277
adopted because of the example of the novelist Luigi da Porto, who, P P g
pretending to investigate old historical records, in reality consulted principally some of the Dante commentators.’
1 Luigi da Porto seems familiar with the commentaries of both Benvenuto da Imola and Francesco da Buti. In the opening paragraph of his novella, he writes:
‘... E avvegnaché io alcune croniche leggendo abbia trovato, come queste due famiglie unite cacciarono Azzo da Esti governator della detta terra, il quale poscia col favore de’ Sam- bonifazi vi ritornd; .. .’
Here he accords perfectly with Benvenuto da Imola, and with no known chronicler.
For the story of the hostility of the two ‘families,’ he doubtless follows Francesco da Buti.
He has also read apparently the Cronica MS. Udinese already referred to.
Onto StaTE UNIVERSITY
THE ICONOGRAPHY AND THE SEQUENCE OF THE AMBULATORY CAPITALS OF CLUNY
By KENNETH JOHN CONANT
Research Associate in Archaeology of the Mediaeval Academy of America
LL extant portions of the Abbey church of Cluny are receiving careful study in connection with the Mediaeval Academy’s project of excavation, and the eight beautiful sculptured capitals from the ambulatory are no exception to this rule. They are well known, ranking among the finest of Romanesque sculptures by uni- versal consent, but they have presented great archaeological problems.
That they were still in their places in 1823 is attested by a litho- graph of the exterior of the apse, dated in that year and republished in SpecuLuM, 11 (1928), 403. The drawing shows the ambulatory roof still in place, though the apse vault had fallen, and probably injured the capitals in its fall. The modern stable, constructed to house the stallions of the government Dépét, covers a large part of the site of the sanctuary. Apparently it was built about 1835, and the capitals were in all probability taken down rather than thrown down when the ruins were demolished to make way for it. Indeed, there is still in use at Cluny an old lewis which fits more recent en- largements in the mediaeval lewis holes.
Five of the capitals bear a stenciled O, which may mean that they were set aside for Doctor Ochier, a local magnate of the day. He had a small lapidary museum in a pavillion near the town hall, and the capitals are said to have formed a part of it. Doctor Ochier's house — none other than the abbatial palace built by Jean de Bourbon — eventually came into the possession of the Town of Cluny, as did his collection. Toward the end of the nineteenth cen- tury the palace was in large part transformed into a museum to receive these and other objects. The fragments from the excava- tions are stored in this building.
The capitals are now set on parts of the original marble shafts on the ground floor of the museum. They bear traces of several unre-
278
KIC FP del
[EJC. ar F Pet R F mens
MeTRESO 1 2 3 4 5
FiGuRE 1
ABBEY CHURCH, ACCORDING TO THE EXCAVATIONS OF 1929
The sequence of the capitals is indicated.
Diversion"\® 7 1@ Corinthian x x
> bur fous x tis Ot vm vn OW v Wom eX! se-@3 5S@:e+ 20°06°~ . ye Wind? Cardi Theo First Prudence jones nal logical Tones Seasons Virtues In @;°e
asPeerMereese &
Figure 2
DIAGRAM oF MunicipaL Museum at CLuUNY, INDICATING LOCATION OF
SHAFTS AND CAPITALS
The shafts are given Roman numerals; the capitals, Arabic numerals.
10
15
Ambulatory Capitals of Cluny 279
lated systems of numbering, so that for convenience a diagram is given here with the shafts numbered 1-xu according to their position in the museum and the capitals numbered 1-8 according to the sequence recently worked out by the present author and set forth in this article. These numbers, as well as those assigned to the transept capitals already published, will be used in the final monograph on the abbey which is to be published by the Mediaeval Academy.
The present arrangement puts the handsomest capitals near the doors and in the centre of the room, and the least broken face to- ward the light. There is nothing to show that any other idea gov- erned the placing of the pieces. There is no known indication on the capitals or in the records of the original grouping, and since the capitals have been moved at least three times and rotated once (by Doctor Pouzet), nothing in their present arrangement can be relied on as evidence for the original arrangement.
The same may be said of the shafts. Number vui still retains an iron dowel on top; shaft rv, the smallest in diameter, was probably near the top of acolumn. The size of the shafts varies from 1’-6;" (0 m. 46) to 1’-7;2" (0 m. 503) in diameter. There are in the museum five thin drums of limestone measuring from 1-7" (0 m. 49) to 1'-7;,” (0 m. 505) in diameter, with cavetto and astragal carved on them, which possibly came immediately under the capitals. One of these (now a misfit on shaft 1) seems to have been painted to match the marble; the profiles correspond to those of the typical engaged columns of the church.
It is said that six of the shafts in the ambulatory were of marble, brought from Rome, and this is entirely probable. They were probably the sumptuosas depolitiones against which St Bernard in- veighed in his Apology to Guillaume de St-Thierry [SpEcULUM, v (1930), 88]. The marble shafts have lewis-holes in the top, with a double slope in the Greek manner or with a single slope in the Roman fashion. They are unfluted and far from perfect geometrically, which implies a Roman date. Several imperfections are or have been filled with lead and painted. Since the ambulatory columns at Cluny were about twenty diameters in height, twelve ancient col- umns may have been used. Shaft 1v, which is of marble, has a
280 Ambulatory Capitals of Cluny
V-cut incised consecration cross 4 };” (0 m. 12) high, resembling the painted one in the Chapel of St Gabriel. Two of the ambulatory columns were carried out in the local limestone. It is worth noting that the plan of the ambulatory chapels at Cluny naturally calls for ten columns in the apse; eight were used instead, probably to avoid putting as many as four limestone columns with the marble ones. Since the first ambulatory arch at each side was much larger than the others, it is reasonable to suppose that the stouter columns were toward the west, the slenderer ones toward the middle; but the study of the capitals is inconclusive on this point. The easternmost piers hid the westernmost columns in a ordinary view down the nave; perhaps, therefore, the limestone columns, which seem to have measured about 1’-7;," (0 m. 492) in diameter, were put at the ends of the series.
For the placing of the capitals we have no direct evidence. Since each capital may have been in any one of four positions on any one of the eight shafts, the number of possible arrangements is very large. The arrangement finally arrived at is plausible rather than certain, for it rests on a tissue of probabilities, but the study was made under the best possible conditions. In the summer of 1929 two sets of casts of the capitals were made by workmen from the Trocadéro Museum, and one of these was sent to Cambridge, the expenses being met by a grant from the John Simon Guggenheim Memorial Founda- tion and certain private funds. The casts were set up in the beauti- fully lighted central court of the Fogg Museum with the original intervals and radius as recovered in the excavations of 1929. All of the promising sequences were tried out, and the fact that they could be studied at leisure under practically the original conditions of space and lighting was of immeasurable assistance.
The following indications were taken into account in establishing the sequence:
1. Owing to the plan of the apse, two of the capitals were set diagonally to the main axis of the church. Therefore it is natural to put Capital 1 (Corinthian) which has its abacus elaborately carved on two adjacent sides, and Capital 8 (the Last Tones) which has its inscriptions across the diagonals, at the ends of the series. The more
Ambulatory Capitals of Cluny 281
elaborately carved of the two adjacent sides of the abacus of Capi- tal 1 faces toward the altar if the capital is placed at the north, and this puts the most broken side toward the apse, where it was most likely to be damaged by falling masonry. A volute of Capital 8 was found in the excavations of the south side of the sanctuary, and it is therefore natural to put this capital on the south side.
11. Related monuments suggest an alternation between Corinthi- anesque and freer forms in the sequence. It is therefore reasonable to put the capitals with mandorlas in the second, fourth, and sixth places. The scale and design of Capital 8 is such that it does not ap- pear to advantage except at the end of the series. The handsomest arrangement of the other capitals places the one with the angular mandorla (the Theological Virtues) as number 4, midway between the capitals with almond-shaped mandorlas.
1. Since the Capital of the Last Tones (8) comes at the southern extremity, it is natural to expect the Capital of the First Tones (2) opposite, suggesting the responses of a monks’ choir. The capital has been found to give a most satisfactory effect when placed second, where it is the first figured capital on the northern side. An excel- lent effect is obtained from all sides if the First Tone is set looking toward the sanctuary. The lines compose beautifully with the rest of the series. The Eight Tones were doubtless included in the icono- graphical scheme to express the love and high cultivation of music at Cluny, and to symbolize the richness of God’s praise which was offered there.
tv. A plausible choice for third place is the Corinthianesque capital (3), with an unbroken figure variously identified as a ‘Wind’ or ‘Harvester’ on account of a bellows-shaped attribute of some wicker-like material. The subject of this capital may never be satis- factorily settled because the remaining figures are so badly broken. A crack across the lower part of the capital suggests that it was damaged by structural disorder in the apse. Capital 4 (the Theolog- ical Virtues) has a larger crack across the middle, and it is quite reasonable to put the two side by side. The most satisfactory ar- rangement of the figures places the most broken side toward the
282 Ambulatory Capitals of Cluny
altar, and the intact corner adjacent to an unbroken rear corner of Capital 4.
v. Capital 4, with its four angular mandorlas, is the smallest of all in lower diameter, and therefore, on account of the probable ar- rangement of the columns suggested above, it should be put in the middle part of the series. Its abacus is more elaborately worked on one side than on the other three, which would seem to indicate a capital looking almost directly toward the nave. A defect in the stone mars a volute on the opposite side. This evidence concurs with the impression gained on purely esthetic grounds and fortifies the reasoning which places the Corinthian Capital (1) at the begin- ning of the series.
The iconography of Capital 4, long enigmatic, has now received its solution. The first figure, which holds an open coffer, is carved under the elaborate side of the abacus, and thus doubtless faced the altar; its iconographic type is used in Gothic times for Largesse, and here doubtless stands for Charity. The succeeding face on the right has a kneeling figure facing east with the hands in the attitude for receiving a consecrated Host, and so represents Faith. The flowered staff and the heart held by the next figure are both used in later times to denote Hope. Thus the three Theological Virtues occur on this capital just north of the axis of the apse, with ‘the greatest of these,’ Charity, looking down into the sanctuary. The remaining figure is almost completely missing. It had a graceful balancing at- titude, with a scale-pan or some very similar object suspended from each of the outstretched hands. Thus it is likely that Justice, one of the Cardinal Virtues, was chosen to fill out the series of Theological Virtues on Capital 4. The Abbot of Cluny had the right of justice in the town, which received its charter from St Hugh.
vi. Iconographically, esthetically, and theoretically, the Capital of the Four Rivers of Paradise is best placed fifth in line, beside the Capital of the Theological Virtues (4). First of all, it is a common- place that the Four Rivers are the figures of the Four Cardinal Virtues. Thus the two types of virtue are brought together on op- posite sides of the main axis. As with Capital 4, the abacus of Capi- tal 5 is worked elaborately on only one side, and that is the most
Ambulatory Capitals of Cluny 283
damaged side, evidently the front of the capital. The back has de- fects in the stone, the sculpture is not undercut as elsewhere, and the figures are not completed — one of them lacks the eyes. On the contrary, the beautiful vine under the elaborately carved abacus- face is most appropriate for the side turned toward the altar. The type of the capital as a whole is sufficiently Corinthianesque to make it take its place gracefully in the alternation, but sufficiently free to avoid any suggestion of rigidity in the alternating scheme. It makes a very satisfying pendant to Capital 4 and takes its place in the sequence most beautifully.
M. Bréhier remarks that the subject of the capital was probably suggested by a letter of Pietro Damiani to St Hugh, of which the monks of Cluny were justly proud. It is given at column 374 of the Patrologia Latina, volume cxutv. It refers to the monastery as to the Four Rivers of Paradise. As will be seen, this capital is the icono- graphic key of the middle group of capitals.
vir. By far the most satisfactory place for the remaining man- dorla capital (6) is that adjoining the Capital of the Four Rivers (5). It thus forms a symmetry with Capital 2 (the First Tones) about the angular mandorla of Capital 4 (the Theological Virtues). This sym- metry is perforce unrelated to the axis of the apse, which relieves it of an obviousness that is only too familiar in modern work. The effect is excellent. The capital has four beautiful female figures, three of them named by carved inscriptions on the respective man- dorlas, Prudentia, #stas, Ver. The fourth figure has been called Grammar and Justice. The fourth inscription was a painted one; it had escaped notice, but was discovered in 1928 and after deciphering copied at full size in 1929, by the present writer.
In fact we have a second manifestation of Prudence — the warner, besides the warrior Prudence already familiar. The figures are best arranged from all points of view when the warrior Pru- dence is set toward the altar; and this is the most broken side of the capital.
The importance of this capital is considerable. The painted in- scription is not different epigraphically from the others. Perhaps all were painted to begin with. If so, this one may have been left un-
284 Ambulatory Capitals of Cluny
cut on account of haste to get the capitals into place. The unfinished back of Capital 4 suggests as much. As was shown in the preceding article, the apse was not constructed until the carven capitals were in their places. Again, the fact that there is a painted inscription points to a calligrapher as the author of the lettering. Since manu- script lettering is generally ahead of carved lettering, the advanced character of the epigraphy of the capitals is explained. Both of these observations strengthen the thesis of the early dating (1088- 1095) of the capitals. A further observation is that the presence of
Figure 3 PAINTED INSCRIPTION ON PRUDENCE-SEASONS CAPITAL
Dat nos monendum Prudentia qid sit agendum.
two Prudences overthrows the too subtle ideologies of those who as- sume the basis of the iconography of the capitals to be mystical sequences of four, such as the Four Temperaments, the Four Winds, the Four Seasons, the Four Liberal Arts, the Four Elements, the Four Ages, the Four Hours, and so forth. The scheme is simpler than that: the four Cardinal Virtues are represented on one capital, the three Theological Virtues on another. One of the Cardinal Virtues is selected for further illustration to fill out the capital of the Theological Virtues (4). Another is selected for further illustration on Capital 6. The double illustration is not illogical, for there is, ac- cording to Professor Panofsky, a Prudence sequence. Here it is
Ambulatory Capitals of Cluny 285
represented by two subjects, just as two subjects are used to repre- sent the sequence of the Seasons, or in a larger sense, Nature.
vi. Thus the figures in the mandorlas and those with attributes or accompanied by inscriptions, fall into three groups: a central one devoted to the Virtues of man, and of the monastery in particular, filled out by a group devoted to Nature, set between two capitals symbolizing divine praises. It is to be noted that the arrangement here is not strictly symmetrical, and that the idea is a relatively simple one. Those who are inclined to date the capitals in the twelfth century naturally look for a subtler and more sophisticated iconography than that of the eleventh century. Here the early dat- ing is once more fortified by the results of the new studies.
ix. It may very well be that Capital 7, which remains to be stu- died, and Capital 3 are simply genre. For Capital 3 (‘Harvester’ [?], ‘Wind ’[?]) this cannot be determined because only one figure re- mains complete. On Capital 7 considerable thought has led to very meagre results. The little man with a furry glove is called Winter, in spite of the fact that the right half of his body is naked, and other conjectures are no more plausible. Yet a good case can be made for the capital as devoted to the sports and diversions. The gloved man thus becomes a boxer or ball-player, robed to make him pre- sentable in church, or to represent him as a cleric. The students of the American seminary in Rome play baseball in cassocks, but mediaeval athletes were almost unclad. The charming nude figure so gracefully twined with leaves becomes a bather or a swimmer; the contorted headless figure, a thrower, bowler, or tumbler; and the simple erect figure, a debater or one walking in meditation. The capital looks best with the first two figures mentioned turned to- ward the altar, but the figures are of minor importance. It is proba- bly quite just to say that the designer of the group of capitals amused himself with a bit of antiquarianism in Capital 1 (Corinthian) and varied the Corinthian theme by the little figures among the leaves in Capitals 3 and 7, while preserving the type as a foil to his more solemn allegorical capitals.
x. The study proceeded to this result purely on the basis of internal evidence, without bringing the actual text of Pietro Damiani
286 Ambulatory Capitals of Cluny
into play. The passage is as follows, referring to what the great churchman had found on his visit to the Monastery of Cluny:
Uidi siquidem paradisum quatuor Euangeliorum fluentis irriguum, imo totidem spiritualium riuis exuberare uirtutum: uidi hortum deliciarum diuersas rosarum ac liliorum gratias germinantem, et mellifluas aromatum ac pigmentorum fragiantias suauiter redolentem, ut de illo uero ualeat Deus omnipotens dicere: Ecce odor filii mei sicut odor agri pleni, cui bene- dixit Dominus. — Migne, Patrologia Latina, cxu1Vv, col. 374.
It is evident at once that the middle group of capitals is an ad- mirable translation into stone of the spirit, the symbolism, and the letter of this text. The rivers and the virtues are there. The flowers and sweet herbs are the beautiful fleurons and other leafage of the capitals; their fragrance is referred to in an inscription of Capital 6 — Uer primos flores primos producit odores. The choice of the seasons Spring and Summer comes naturally from the refer- ence to growing things. The ensemble represents the Abbey itself and its spiritual beauties. One rarely finds so apt and lovely an expression in line and surface of such beautiful thoughts and words.
The letter was addressed to St Hugh of Semur and had a unique significance for him as a benediction on his life’s work. It is more natural that he should have used it as the basis of an iconographical scheme than that a successor, a generation later, should have done so. The text therefore reinforces the idea that the iconography was of a free and early type, fortifies the interpretation arrived at inde- pendently from internal evidence, and strengthens the thesis of early dating for the capitals.
x1. Capital 8 (the Last Tones) seems to be placed by its inscrip- tions; at any rate it is sensible to have the first quarter of the in- scription on the diagonal toward the nave, and this gives a very acceptable arrangement of the figures as seen from both apse and ambulatory.
This capital is a Cinderella. It is planned to restore and complete the designs of the series of capitals, using the Fogg Museum casts; and Capital 8 was chosen as the first. Copies of the single extant volute were attached to the broken corners, and the missing abacus was supplied by one studied from the back of Capital 5. These ad-
The faces re Presented »
Capital 1. Capital 2.
Corinthian The First Four Tones of Plainsong The W
PLATE Ia SEQUENCE OF THE AMBULATORY CAPITALS OF THE ABBEY OF CLUNY
robably looked toward the Altar. The scale of the re production is about 1:17, the same as that used for the transe pt capitals in SPECI
Capital 4. Capital 5 Capital 6.
The Theological Virtues and Justice The Four Rivers of Paradise Prudence, Summer, Spr i
Casts OF THE CLUNY CAPITALS ARRANGED IN SEQUENCE IN THE CourT OF THE FoGcG Museum, CAMBRIDGE The original intervals and radius were used in placing the casts.
Rocce
transept capitals in SpecuLUM, IV, 302 ff.
Capital 6 Capital 7.
Prudence, Summe:, Spring Sports and Diversions (?) or genre
a, CAMBRIDGE
Boecker phot
Loury phot
Capital 8.
The Last Four Tones of Plainsong
wr genre
Ambulatory Capitals of Cluny 287
ditions transformed the ineffective broken design into one of amaz- ing force, originality, and beauty. The stems of the volutes, as seen from below, form behind the figures a clever variant of the man- dorlas of Capitals 2, 4, and 6. This continues the alternation and serves also to distinguish Capital 8 from the genre capitals num- bered 3 and 7.
It is worth noting, moreover, that the swing of the volutes is sur- prisingly like that on several figured capitals from Frémista, Leén, and Iguacel, published by Mr Porter in Spanish Romanesque Sculp- ture (Paris: Pegasus, 1928; plates 24, 25, 26, 27, 30, 31). Since these sculptures are older than, or contemporary with, the apse of Cluny, they may be studied with profit both by those who do, and those who do not believe that the Cluny capitals were sculptured before 1095. The style is not the same, but the idea that figures in high re- lief were not used on capitals before 1100 is contradicted by these examples. Mr Whitehill’s important and still unpublished dis- coveries at Santo Domingo de Silos tend to clarify the relation of that important work to Cluny. The most acceptable precursor for the existing transept design at Cluny is to be found in the Castle Chapel at Loarre in Spain. Clearly the connection between Cluny
and Spain was by no means a tenuous one. As a matter of fact Spain can no more be left out of account than the north, from which came Hézelon of Liége, architect of the church. Writers in general have failed to recognize the international and synthesizing character of the great monument at Cluny, and French writers in particular, by forcing the capitals into a regular system of French development, have mistaken both their date and their significance.
Harvarp UNIVERSITY
THE TEXTUAL CRITICISM OF MEDIAEVAL LATIN POETS
Br WALTER BRADBURY SEDGWICK
T seems a pity that the student of Mediaeval Latin should still be
in the plight of the classical scholars of the Renaissance, and constantly have to make his own text as he reads. Much unneces- sary expenditure of time (and temper) is caused by the endeavour to make sense of an obviously corrupt text. Editors are not altogether to blame — especially in the case of an editio princeps: they quite properly assume that it is their first duty to publish the results of their researches as early as possible. Moreover, there is always the risk of obliterating an important manuscript reading by a super- ficial and obvious ‘correction,’ while even the errors of manuscripts may of course be instructive.
But the fact remains that very often editors of mediaeval texts are little interested in textual questions, being palaeographers, historians, or anything rather than textual critics, while their readers are usually too occupied with their own studies to devote much time to textual questions. Hence it is that texts are too often printed and reprinted (sometimes with a fairly elaborate critical apparatus) with unintelligible, ungrammatical, and unmetrical readings by the score — and never an obelus to warn the reader. It really is too bad that Hart’s two Pyramus poems, published in 1889, should have been twice reprinted by eminent scholars like Edmond Faral and Paul Lehmann (in 1913 and 1927) without any attempt to remove or even point out the grossest errors.
Scholars who are trained textual critics usually leave mediaeval texts severely alone, but there are exceptions. For instance, Haupt more than once exercised his critical skill on mediaeval poets (some- times corrupting them by too rigid adherence to Classical usage), while Havet applied his acumen to the newly discovered poet Mahieu de Boulogne.
Pending the arrival of new Havets and Haupts much might be done for mediaeval texts, especially the poets, if scholars would print
288
The Textual Criticism of Mediaeval Latin Poets 289
without unnecessary padding such corrections as are bound to occur to them if they are bent on extracting sense out of the texts they read. The problem is somewhat different from that offered by Classical texts, as the manuscripts are usually not far removed from the auto- graph: on this see L. Havet, Manuel de Critique Verbale (Paris: Hachette, 1911), §§ 654 ff. As Havet says, ‘fautes banales’ here ac- quire some importance for method: he quotes as examples the inter- change in a manuscript of about 1300 on n and u; im, mi, nu, un; e,r, t; e and o. To these I may add a few examples from my own
observation.
Confusion of letters: s < f: passim, e. g. Wright’s Anglo-Lat. Sa- tirical Poets, 11, 251, 1. 5 Salerna for Falerna; satis for fontis, Lehmann (cited below) p. 60, 1. 344.
el <d:clarus for durus, Henr. Sept. 482 (and Poetria Noua 802?); clara for data, Ps.-Hildebert, Excid. Troi., 1. 1. h <n: hominis for nominis, Wright, 1, 251, 8. ¢ <t: conto for toto (or tanto) Architr., p. 244, 1. 15, Wright; sic for sit, Wright 11, 168, 6; et quis for ecquis, Wright, 1, 206, 4 up (i.e. from the bottom); teli for celi, Wright, u, 208, 3 up; Orcus for (h)ortus, Faral, Arts Poét., p. 128, v. 54; intus for incus, Laborintus 917. esorem for esocem, Brunellus, p. 108, 1. 4, Wright. metet for meret, Adolf. Vien. 306; ferte for ferre, Wright, u, 171; censere for censete, Wright, 11, 214, 15; secura for secuta, Anticl. 1, 419; arcere for arcete, Wright, 11, 128, 23; clara for data above. r <v(u): wocaret for uocavit? Wright, 11, 167, 8; rewtuescere for re- utrescere, ibid., 11, 216; rerum for uerum, tbid., 11, 299, 10 (cf. H. Sept. 670?) t <v (u): teste for ueste, Lehmann, p. 61, 1. 390. n <v (u): moueant for moneant, ibid., p. 63, 1. 4. a <e: passim, e. g. Wright, u, 167, 5 up, qua for que (quae); Brunel., p. 39, 5 up (Wright) amica for amice. u <a: furius for furias, Wright, 11, 65, 2; sociam for socium, tbid., p. 943.
290 The Textual Criticism of Mediaeval Latin Poets
Transposition of letters: ampla for palma, Faral, op. cit., p. 156, 18, v.2; moderat for mordeat, Anticlaud. 1, 339; palmas for lampas ? Adolf. Vien. 326; uenerunt for neuerunt, Brunel. 66. 3, 97.10 up (Wright): cf. Laborintus 608; inueniat for ni ueniat, Lehmann, p. 57, 1. 214.
Pronunciation: soft c as s: sensus for census, Wright, 11, 239, 5 up; sensura for censura, Wright, 1, 231, ult.; census for sensus, Anticl. 1, 144; seu for ceu, Brunel. 68, 8; sit for scit, Wright, m1, 204, 8 up; recipissent for resipiscent, ibid., 11,207, 3 up. Final d and ¢ confused: e. g. inquid (passim); quod for quot, Wright, 11, 238, 8.
Classical reminiscences sometimes help: e. g. in Anticl. ii, 304, l. 11, the imitation of Sidonius, Ep. iv, 3 shows we must read Sollius for Soldius. Ibid., 1, 288, 1. 2 up, is a reminiscence of Ov., Met. ii. 14, hence we must read diuersa for aduersa. Lehmann, p. 57, |. 208; ambrosiam may be restored for umbrosam from Ov., Met. iv.215.
In attempting to emend a text of the twelfth to fourteenth cen- turies, we must first decide on general grounds what to expect. On the whole poets of the twelfth century may be expected to be well trained in the mechanical part of their art, and to show the influence of the schools of Northern France, the teaching of which is repres- ented by Faral’s Arts Poétiques: of these Matthew of Vendéme, in theory and practice, is of fundamental importance. So thoroughly was he studied that many poems (e. g. the Alda of William of Blois) are in style indistinguishable from his own work. In this school of poetry, then, we may expect perfect fluency and mastery of versifica- tion, ingenuity of expression, sequence of thought and coherence of language. False quantities are extremely rare, though on the model of Christian poets some words have unclassical scansion, and in cer- tain cases the prosody taught in the Middle Ages differs from that of classical times. A verse which is simply incorrect is almost, if not quite, unheard of.
In the thirteenth and fourteenth centuries a decadence set in, and much incompetent work was produced: teaching seems to have de- teriorated, even largely to have ceased. We cannot always expect the smoothness and ingenuity of the twelfth century. German poets in particular, who now become active, seem to me often clumsy and
The Textual Criticism of Mediaeval Latin Poets 291
awkward. False quantities become commoner (though still rare), but a verse which absolutely will not scan is still almost incon- ceivable.
By about 1300 the collapse is complete, and we may look for al- most any nonsense, though even now metrically faulty lines are very rare. In Lehmann’s book we see specimens of all stages. I do not suppose Matthew and Baudri (first half of the twelfth century) were much better poets than the authors of the De Somnio and the poem (No. v) about Venus, but they knew their trade, which the later pair (ca. 1300?) emphatically did not. Between these extremes stand the remaining three Pyramus poems (xm cent?): a and 6 have thyme, c not; c is probably earlier — it stands nearer to the tradition of the schools (cf., for example, Il. 58, 69).
For the rest the ordinary rules of textual criticism must be applied, and in particular, causes of corruption carefully classified, as so many texts depend on a single untrustworthy manuscript; e.g. in the manu- script of Matthew’s Pyramus (Lehmann, pp. 31 ff.), we note the tendency to mechanical repetition of syllables (see the notes on ll. 93, 108, 132): this helps to correct ]. 132, and denotes corruption in ll. 1, 2, 13, 74, 98, 102, 114; cf. uultus in Il. 138, 141, 142; mortificare in |. 148; compactente loco in 172 may be an unconscious reminiscence of ]. 34. The intelligence of the scribe of H in Lehmann, pp. 35ff. is shown by his writing en for est in Il. 31, 37, 50, 79.
Finally if editors would pay more attention to punctuation, the reader’s task would be considerably lightened. Wright is a great sinner in this respect, but Lehmann is nearly as bad. I deal below with a few of the more difficult cases, but I pass over hundreds in silence.
Wright’s Anglo-Latin Satirical Poets of the Twelfth Century
(Rolls Series, 1872)
The following emendations are offered for what they are worth, with the object of saving the readers’ time and contributing towards amore readable text. The texts vary greatly in purity: some few are hopeless, others require few corrections. I may perhaps be allowed
292 The Textual Criticism of Mediaeval Latin Poets
to say that the result of a somewhat minute reading of these authors has been to increase my respect for their intellectual powers, while extinguishing what little I had for their poetic inspiration. In al- most every text we are justified in expecting good Mediaeval Latin, good sense, and logical sequence: it is unjust to our authors to think them capable of writing mere rubbish; with the exception of men like Serlo they were highly cultivated stylists.
I have not as a rule troubled to note all the obvious changes which the reader can make currente calamo (amat for amet, etc.), nor changes of punctuation where the sense was not seriously affected, nor indi- cated the very numerous cases where I prefer rejected readings: this effects a considerable saving of space; for the same purpose I simply print Wright’s reading followed by my own conjecture, in cases of doubt followed by one or two marks of interrogation, without com- ment.
VOLUME I Speculum Stultorum (or Brunellus)
The variants of manuscript B are very often right, perhaps nearly always, apart from lapsus calamt, e.g. in the notes pp. 74. 10, 75. 1, 76. 1, 2, (6?), 77. 2, 5, 78. 4, 79. 2, 4, 6, 9, and especially p. 80. Page 11, 1. 11 mdgis: nugis. 16.3 oculis: maculis? 28.3 comma after minus, semicolon after habemus. 29. 10 note uespes = vespae: cf. 1.19; B twice spells it waspes (an Anglicism?). 29.16 quaeque tuta minus: quoque minus tuta, or, better, quae, quo tuta minus [est, eo] fortius ... 33. 11 up, lucis: lut(e)ris, with p. 33.6 up, flamine: stamine, with B, P. 39.5 up, amica: amice. 40 ult. num: sum. 42.8 up, haec est noua regula. 44.5 up, mercarum: marcarum. Pp. 43-44. I cannot under- stand why dédecus is confined to these pages. Are they interpolated? Elsewhere the word is always rightly scanned. In 43.2 up we could read dedecus ut breue sit: 44.7 dedecus hoc sit, with P: 44.9 dedecus est: but 44.3 I cannot emend.? 56.15 awe: auet?? 57.3 onerant <que> cibaria mensas. 58.3 potus (toti): poti? 60.12 Iam[que] bobus . .
1 In Architrenius, vi, viii (p. 332.6 up, in Wright), stamen tamen Atropos aeui rumpil, the editions vary between flamen aeui and stamen alui—a notable instance of mediaeval
editing. 2 Perhaps in dominum nostrum (papam is a gloss) si dedecus usque redundat.
The Textual Criticism of Mediaeval Latin Poets 293
uergit: uertit. 65.10 up, petito: petitum. 66.13 longo mea stamina Parcae uenerunt: longe . . . neuerunt (cf. 97. 10 up). 67.6 et: ut. 67.7 cur: cum. 67.9 huic: hinc. 68.8 up, seu: ceu. 74.1 conduzxere: condixere (P). 75.14 up, inde: unda?? 87.11 laxant: laxent. 88, n.5: B right, if venerunt is changed to neuerunt (cf. 66.13). 99.4 luat (luit, B): linit (from linire), ‘defiles.’ 99.13 auarum: auari, subject of delirant. 103.12 inarmatas. 105.8. B right, with comma only after mihi. 108.6 anipitrem: ancipitrem, cf. p. 110, n.8. 108.4. esorem: esocem (Pliny). 108.14 mozx: hora; quod est: breuis?? cf. hora breuis, 1. 12. 108 ult. haec: hic? 113.3 up, canem: perhaps cadum. 117.9 auidus: auibus. (in 10, P is right). 118, 11 up, proderés ??. 120.11 nolunt: nolint. 123.7 uerna[que]. 126.8 redderit: redderet. 127.2 up, uigent: possibly rident (or uernant). 131.6 up, omni: esse, with B. 131.7 omni: omine? 132.10 effractus: effetus? 134.6 aliis: aliter. 137.14 up, tu autem, sign of farewell, cf. Voigt on Ysengrim, 1, 927. 137.13 up, Dryano: Dryane.
ARCHITRENIUS
The text is corrupt, and complicated by Wright’s bad punctuation. P. 240 |. 9. read wenativa morae, uix inceptura quod ... (in |. 7. del. comma after nutricia). 240.4 up, del. comma. 240.3 up, fame = famae. 240 ult., del. full stop. 241. 6ff. ‘not tempted by glory which crowns toil, nor by rewards, by which toil learns to become pleasant’ 241.11 tristitiis: tristities. 4 up, uersum: uersu. 242.11 multum: mutum? 243.11 up, seruum: feruens? 244.4 the full stop should be after tulit. 244.12 up: the second hic should be haec? 245.4 up, floreri: florere. 245.14 (down) weris: cineris. 246. 6f. read ‘et, quo utdet Anglia, reddi poscit inoccidui commissum luminis usum’: asks that the use of the unfailing light, by which England sees, be restored: usus commissus)(mancipio datus. 247.8 up, momenta: monumenta? ? 249f. The information is mostly from Pliny, H. N. vii, but whence the queer story about the mother of the Gracchi? 249.3. coniectura: coniunctura or commiatura. 250, 8 up, trata Bythia should be genitive. 250 ult. miseris lethos: misero fletu. 252.2 euicet: ericit. 252.6 circumeat: circummeat. 252.9 planos: planis. 252 ult. possibly locum for loci, and comma before coruscat, not after.
294 The Textual Criticism of Mediaeval Latin Poets
253.8 del. full stop. 253.13 juncto: justo? 255.13 up, del. full stop. 256 ult. auarae: auare; comma after addit. 258.7 comma before, not after dulce. (As I could only borrow these volumes for a limited period, I wasted no more time on this dull, turgid, and obscure poem). VOLUME II Godefredi Epigrammata
These are singularly feeble: several of them are beyond me, but I offer one or two suggestions. Ep. vi.1 /aus must stand for the vocative of a proper name, e.g. Lause. 20.2 ita uelis: I caue sis follows the ductus litterarum. 21 dum . . . uenor all belongs to the protasis. 34.2 del. commas, and put one after aliis. 40 is unintelligible. Per- haps we should keep the reading of C, changing haud to hoc, 62.1 nonius humor: we expect roscidus. 65.2 furius: furtas. 91.1 semicolon at end. 94.2 soctam: socitum. 115.4 read ludum defendis? uertere, liber eris (pun on liber). 156.2 et: ut. 157 refers to Publilius Syrus 64 (see Skeat’s note to Chaucer’s Melibeus, C.T., B 3049).
Henrici Epigrammata
P. 163. Pref. 1.7 ststante: sis tantae. verse 5 refrigerent: refigerent. verse 13, comma after wersus, not before. P. 164.12 sit: fit. 165.3 pergamenta: purgamenta. 165.13 up, hi: hic. 166, ad Zetam |. 4 quaequae: quemquam. 167.7 testa: caenoque or turbaque?? 167.8 uocaret: uocauit? 167.11 up, faustu: fastu? 167.0 up, somnia qua: somnia; quae. 168.3 put the semicolon after clerus and continue the speech to |. 6. Penultimate line of poem male for mala. Last line iustis: iustus et. P. 168, Epitaph, 1.6 sic: sit? P. 171, de Auaritia: read quisquis auare uixit, without comma. de Zoilo, |. 1 hominis: nominis. L. 2 ferte: ferre. de Cont. Visib 1.3 mecum: tecum. 7 quaeras: quaeris. 9 one question mark only, after miror. p. 172.2 the comma should come after rerum. 13 punctuate, quis uilia? 19 del. et after decet. 27 parate pingues, 1.e. ‘make your owners fat’; del. comma. P. 173, 6-9 is the advice of the soul (in 9 del. comma after nobilitas), 11-16 of the body (in 15 semicolon before and after duplicasti). 10 up; the second raperes should be rapis.
The Textual Criticism of Mediaeval Latin Poets 295
ALEXANDER NeckaM: de Vita Monachorum
P. 177.9 up, aestimat: aestuat (illa). 179.13 speculo: specula. 180.7 up, permanet: praeterit? 182.2 si: sed. 182.13 sic: dic. 182.19 est: esse. 183.8 substitute comma for full stop. 183.11 forma: fama. 184.11 up, possunt: possint. 185.14 ingenium: ingenuum. 186.6 comma after mtror, not carnis. 187.10 up, namque: nam quae. 188.7 nec: non. 188.9 facte: facies. 188.15 intra: intrat. 188.16 nolo: uolo. 188.18, add et after monita. 188.23 arcere: arcete. 189.3 a syllable is lacking, perhaps sibi after quantum. 189.7 up, aewum: foedum? 189.5 up lege: nece? 189 ult. et quis: ecquis. 190.5 est: erit. 190.13f. add semicolon after addere; quicquam: quicquid (= ‘every- thing’); nouwm: noutt. 191.2, insert question mark after ad quid. 191.5, add -que after agricolae. 191.4 up similis: simili. 192.3 si .... conspicis: sic... .conspicit ?? 192.13 up del. comma. 193.10 sic: uiz. 193.13 magnus: Magnus. 194.2 up tot arborum: arbustorum ? (tot from an interlinear correction of arborum). 195.17 up haeret: hebet. 196.10 up del. comma and put comma for full stop. 196.9 up manent: manet. 198.6 up, afficiat: afficiet. 198.4 up, -que: quoque. 199.2 redonatae: retro natae?? 200.5 up, tacitus: tectos. 200.6 up, si: sed. Gualonis Inuectio. 202.4 for first nomtna, read crimina, 203.6 up acuunt, acuit. 203.4 up contigit: colligit. 204.20 up fit: sit. 204.15 up Odardus: tardus. 204.11 up dura: durae. 204.8 up sit: scit. 205.12 refutant: recusant? 205.16 colant: celant. 205.17 semicolon after minus. In 205.11 up, as at 205.6 up and 207 massa seems to mean ‘bulk.’ 205.8 up quis: quid. 206.2 qui propria: transpose. 206.11 offerrt: offer et. 206.12 satagat = ‘satisfy’: (sat agat?), triformae, triforme 206.15 provectus: profectus. 206.22 illum: illud. 206.4 up estis: escis. 207.3 sunt bona: transpose. 207.12 up nam: nunc. 207.3 up recipissent: resipiscent (interesting for pronun- ciation).
Serlo
P. 208.1 qui: quam. 208.4 iuvant; utuunt. 208.6 omnis: omnes; comprimet: comprimat. 208.11 up, parens: patris. 208.3 up Teli: coeli. 209.18 up accendere: attendere. 209.10 up sancti: sacra? 210.1 tales... . monachales: talis ....monachalis. 210.2 mark of
296 The Textual Criticism of Mediaeval Latin Poets
exclamation after pestis: insert haec after facit. 210.34 transpose qui and quod. 210.5 the reference is to the black habit of monks. 210.9 up studuerunt: statuerunt (the whole is irony). 211.5 up et: ut. 211 ult. accipiunt: accipiant. 212.2 put the comma after suarum. 212.4 comma after erramus, colon after tpsi. The last six lines are a different poem (epitaph).
P. 213. Anon. 10 up tera talem: timeat tales? 213.9 up donetur: donet te. 213.3 up, del. full stop. 214.15 censere: censete. 214.12 up burburus immitis = Marbod., pp. 86 and 92 (ed. Ropartz, Rennes, n.d). 214.3 up, add fur after petitt (or meruit?). The reference is to the crucifixion. 215.17 up, semicolon after /iceat: comma after in- quinat, 216.7 up reutuescere: reuirescere (ficos = ‘piles,’ Mart. i. 66). 217.3 mélestia, as Marbod., p. 92 (ed. cit.), J. Garl. Syn. 328 (ed. Leyser) — from méles. 217.15 up possibly at should be inserted before sine: but facit seems corrupt (sancit?).
Hueco SorovaGensis: Versus.
P. 222.3 unicuique: sua cuique 224.10 up, mentitur: metitur. 226.5 perueniat: praeueniat. 226.11 up, cathedrans: cathedra, or — as? 226.6 up, mouens: monens. est 228, 11, del. comma after dives, semi- colon after egents.
Anon. p. 231 ult. sensura: censura.
Serlo: p. 232.8 up, mundi: mundus? 234.17 se: st 234.9 up, pura: cura. 234.8 up, de: et. 235.12 paxillum: pauxillum. 236.8 hoc: haec. 236.21 luditis: ludis. 236.5 up, cordi (end of line): sordi. 237.10 laxat: laxas; del. comma. 237.16 ditat (at end): uitat. 237.2 up, nolli: molli. The missing portions may be huic .. . inmiti... hune. 238.3 quod: quot. 238.8 del. comma. 238.4 up, pedis: pede. (comma before regni, not after. 239.7 ne quis: nequis 239.8 del. comma after quantum (uoluens is strange: ualeas would suit the sense). 239.14 up; genitoris: genitor. 12 and 9 up. del. full stops. 6 up, full stop at end. 5 up, sensu: censu. 240, 1-2 semic. after egens, query after deteriores. 240.10 ne: et? (del. comma after ago.) 240.12 haec: has? 241.4 up, precatio: praedatio? 243.6 properautt: properant. 243.11 recessus: census?? (‘wealth’). 244.9, del. ef. 244.13 up,
The Textual Criticism of Mediaeval Latin Poets 297
postrema sonare: post rem resonare 245.4 up; lam nova? 246.15 qua: quam. 246.9 up, semic. after sordida. 246.5 up, manserunt: reman- serunt (or quae for qua). 247.10 up, non mihi: nam mea? 248.11 pugnat is: pugnans. 249.12 hac: hic. 249.12 up, O armis: carminis. 250.14 haec: hoc. 250.3 up, pecte: pectore.
Inuectio in Gillebertum is about the most corrupt text I know: many lines (including the first) are beyond me altogether.
251.3 suus: sitts. 251.5 Salerna: Falerna. 251.7 uernus: uacuus, eam: tam. 251.8 formam: sortem? 251.11 del. full stop; for conterat read concutit. 251.13 praestit: praefert? 251.7 up, taculatori: tocula- tori. 251, 2-4 up, are beyond my powers (in 2 up, for dicel rhyme suggests decem: cf. bis trina above, and bis senos below). 252.7 uero:
Nero. 252.10 quam: qua. 252.14 qualis et nepisti: quae concepisti (no stop at end), ef. J. Bridlington in Wright’s Pol. Poems and Songs (Rolls Series, 1859) 1.47 ‘stercora concepit, peperit quae ventre re- cept.” 252.15 Musa te: te Musa. 252.17 up, muneri: moneri. 252.16 up, sero: scio. 252.15 up, opus crepis: os obstrepit. 252.13 up, del.
full stop. 252.12 up, cogeris: cogens. 252.11 up, uer ima: uis cura. 252.10 up, comma after censura, not aequi: read sorbens nostra et tua? 252.4 up, offendo mando: offis donando?? 252,3 up, f ... : faece? alternorum: alienorum. 253.1 tonitrua. 253.8 sperneres: spernere, comma after patrum. 253.15 potestatis: pietatis. 253.16 movet: tuvat? ? 253.19 mones sic enutritt: monet sic enutrire. 253.11 up, semicolon after uiduamque. 253.6 up, homicida: homicidia. 253.4 up, guies esset, mentem tibi: quaesisset messem mihi. 253 ult. inuasis- set: nunc urxisset.
Anticlaudianus
Bk 1. P. 276, & up, pingunt: fingunt. p. 278, 11 up, excepto: ex- cepta? ? (nom. sing.) 284, 4 f., transfer semicolon to materiae; for neutri read neutra. 285, ult. monet: mouet. 288.2 up, aduersa: diuersa (= Ov., Met. ii.14).
Bk u. P. 296.5 quis: quid. 297.2 quaeritur: queritur. 299.13 functus: fructus 299.10 rerum: uerum, with full stop-at end, not at I. 11. 302.12 and 14 diuidat, uenrat. 302.8 up, fauorem: laborem (question ends here?). 302 ult., del. semicolon. 303.6 up, sense seems to de-
298 The Textual Criticism of Mediaeval Latin Poets
mand quam for quid. 303.4 up, Milo: Myro? 304.11 ut Sodius im- plicat: read Sollius (Sid. Apollinaris, of whom it is an imitation, Ep. iv, 3). 305.14 praescribit: proscribit. 309.8 Dindimus, an inter- locutor in Hugh of St Victor’s dialogues (name from legendary his- tory of Alexander?) 309.15 ramor, is of course rumor: oberret, aberret ?
Bk m1. P. 313.9 up, hispide: hispida. 315.6 up, picturata: pictura. 319.12 cadit: cedit. 319.15 operosae: operose. 322.11 praefecta: praedicta? as in parallel passage 308.14 eleufuga: see commentators on Chaucer’s Troilus, iii. 933 (it is the pons asinorum, not Eucl., i. 47 as Chaucer implies). 320, 4ff. artifices (‘geometers’) is antecedent of qui, tractus . . . pondera is object of claudentes, aera . . . terras of tuentur. 8 ff. comma after orbem, not tractus.
Bk tv. P. 1.4 del. comma after degenerat. 331.10 alli: illis? facies is queer: series?? 332.3 superum should apparently be superam. 333.4 Pyrous, i.e. Pyroeis (Ov., Met. 2.153). 333.3 up, uino: Iuno (obviously the giver of the horse, as Jupiter of the first). 336.7 praedicto iugalt: the genitive seems better. 336.8 minus: nimis. 338.6 nubilia: nubila.
Bk v. P. 349.10 up instantia should be fem. sing. hence read refellit with B, M (Sprcuoum, vol. 11 (1927), p. 341, 1. 5). 359.3 up, nutu: uultu ??
Bk vi. P. 366.11 elata: the variants point to clauata (‘striped’). 369.1 medicinam: medicinae (posse is a noun). 369.13 up, usu specu- lante is abl. abs.; del. comma. 369.14 up, fulgens: fugiens (masc.)?? 370 ult., no pause between Ch. ii and iii. 371.12 up, excepti: exceptum?? 375.10 faciem (que). 376.13 regati: legali? 379.13 con- tempta, i.e. contenta. 379.6 up, exempta: exempla. 379.2 up, Tudith: Dauidae? 381.2 up, manus.
Bk vit. P. 387.1 desertum: deiectum? 387.12 up, nimit: nimio? 387 ult. nostrae [Diones]: perhaps taetrae or the like. 396 ult. alternat: alternant? 396.2 propinqua. 397.5 parit: perit? 398.3 unde: unda (del. comma after gignit).
Bk vit.P. 399.11 ipsa illa: read tsta. 410.13 dat: dant.
Bk rx. P. 414. ult. semicolons after fustes, rapit, siti. 414.7 up, argumenta: hardly armamenta; cf. 420.14 up, 415.11 up, coecus; add: coetus (i.qg. omnis facultas and conciues below). 417.2 B has right
The Textual Criticism of Mediaeval Latin Poets 299
reading. 417.14 pro Natura: pro nata? cf. Ch. v. 1.2; but B, M may well be right (prona tamen). 418.11 up, uarvo: uano. 419.11 up, uulnus: uultus. 420.9 metitur: mentitur.
In the following suggestions I do not claim to have dealt with every difficulty: in the case of Adolphus the text is so corrupt and the author so ignorant that many passages seem quite hopeless. Still it is time something was done towards removing the most flagrant corruptions in mediaeval Latin texts. It is to be hoped that future editors will follow Marigo’s example in indexing the chief peculiarities of their authors.
Henrici Septimellensis Elegia ' ed. Marigo, Draghi: Padua, 1926.
Line 85 cupidas: read rapidas. 136 ablato: perhaps alato, taking remige as ‘oarage’ (after Virgil’s remzgto alarum), cf. 1. 694; but see SPECULUM, 111 (1928), 370. 189 unquam: read tnquam. 314 caluis, edd. calphos: a chess term, usually Alphilus or Alpinus. Read Al- philis, omitting et? 398 insere: perhaps ingere, and reticenda of edd.
408 f. nescts, ferts are incredible: the other really false quantities, e.g. udigti and péra are much more excusable (Il. 242, 838). Perhaps we should read nescia (elision in second half of pent., as at 502), and tu furis atque feris in 409, a very uncertain line (with meque for meue?). 482 clarus: read durus. 493 abest: perhaps aues. 498 uetat: perhaps negat (sc. lex). 502 Fortuna: read Fortunae. 531 multaque: perhaps multane, interrogatively. 654 Marigo glosses tenut . . . omine as filo, but is this possible? perhaps Henry wrote stamine (or nemine): or omine might = ‘rumor,’ ‘leuis aura.’ 670 uerus: perhaps rerum. 681 hardly right. Perhaps cum retulisti, taking retulisti with both unde and cum: ‘Why did you say (I re- member well your saying)?’ 683 hec: hoc 685 ueniamus is incredible and ill-attested: perhaps puro deuenimus, or non puro v. auro (the best MSS have uenimus). 724 ius pia: read impia. 870 adoratus: misprint for odoratus? 908 seque crumenat amor: a very strange
1 See now Karl Strecker’s important article in Studi Medievali 11 (1929) 110-133.
300 =The Textual Criticism of Mediaeval Latin Poets
phrase; I don’t know if deque crumenat is any better; tmesis is com- mon in Henry, and compare deremitat (905). It is barely possible that orbiculat was derived by him from orbus: ‘love blinds and robs you.’
The following changes of punctuation seem to me improvements: 166 ‘pauper .. . tacet’ in inverted commas: ‘If the adage (Ov., Fast. i.218) pauper ubique tacet is true.’ (Henry puns on iacet). 197 put comma before dira, not after. 334 comma after suscipiam. 5438 delete commas: (ignare = ‘ignorantly’). 759 del. comma after est.
The following variants seem to me highly probable:
169 inititum (usual mediaeval scansion). 299 hoc. 232 meritis. 478 aspis was perhaps Henry’s own spelling. 504 limen. 522 non, vel uiz. 583 suffdcat (or sustulit.) 651 Macedum. 714 ingenuam talem (opposed to witiata), 719 nunc (Leyser)restores metre and adds point. 759 duplicat (cf. 1. 529): décuplat has no real parallel in H. 770 ad- uersis (Leyser). 780 turpe is a gloss: read non (cf. Hor. Ep. I, ii. 3). 797 arcum (hardly hamum). 950 read insidie; sit.
Adolfus Viennensis
reprinted (in part) from Leyser by Jakob Ulrich, Lat. Novellistik des Mittelalters, Leipzig, 1906
Line 20 hanc puram pure ne luat (‘defile’ cf. 1. 394) hanc alius. 29 (ex) clamat. 35 varia mulier mala. 58 nec: read nunc. 86 omit cum. 90 (super) exaltando 106 eis hebeat. 129 inuenies: iuvenis. 175 est et. 177 mala: read male. 192 precanti: peccanti. 209 hic uetulaeque graut dicit. 215 the semicolon should come after securus. 216 uolo: read ualeo. 219 delete comma after est (‘eats’). 223 widit. 236 comma after poenarum. 258 uulpida. 274 diem: read die. 277 per- haps omne; dat huic tlla: pronouns as in |. 303 f. 297 uel: read uelut. 299 reseraque. 302 ut patre a summo. 306 metet: read meret (cf. 1. 301). 307 refert (as 1. 273). 309 aspis. 322 haec: sponso. 326 palmis: hardly lampas (propius in 325) 335 scelus et. 351 Tunc uir eam nimium precibus orautt amicis (for the lengthening at fourth ictus cf. H. Sept., 509, 511, etc.). 361 perhaps feruere uagari lena ac turpra (or feruereque; cf. 1. 252: or feruescere lena: amidst A’s erratic uses of et and que it is difficult to come to a decision). 367 (dum) stultus.
The Textual Criticism of Mediaeval Latin Poets 301
P. LeumMann, Pseudo-antike Literatur des Mittelalters (Teubner: Leipzig, 1927)
These notes were finished in December 1927. I now find that Karl Strecker in Hermes, txu, ii (1927), 243 ff. deals minutely with the third Pyramus poem (pp. 52 ff.), and anticipates many of my con- jectures. He also kindly communicates the readings in the other poems marked ‘K. 8.’
I. Pyramus and Thisbe, by Matthew of Vendéme, pp. 31 ff., 1. 1, amoris: erroris. 2 insanus: insanas? (Il. 1, 2, being Leonines, are spurious, K. S$.) 13 dissimilis: discrimine. 17 comma after eget, not sensus (degenerare, ‘belie their nature,’ as 1]. 23, etc., see SPECULUM, 1 (1928), 373. 25 pueritia is scarcely credible in Matthew: perhaps puerilt labitur aeuo (in Venerem puerts iam labitur annus, tus habet in iuvenes, K. S.). 37 cui: qua? 40 comma after, not before, murmura. 62 risus, K. S. 64 f. Matthew’s precept and practice demand stop at end of pentameter. 65 perhaps tgnis uterque furit: sic ledit, leditur, urit.... but it is obscure to me. 69 petit: petunt; no comma after solacia. 73 periussa, etc.: perhaps perit ipsa licencia uocis. 74 rimat: uacat? 77 male: mala. 91 nec: hac. 97 f. del. comma after fidelis, and read femineam (but fraudem is suspicious). 99 witro: uitri (facundia, K.S.). 101 armatis K.S. 102 sanguinis: sanguine. 103 ferinos deuitat, K.S. 114 accelerare: anticipare. 117 Tisbesque. 120 solet propheta: invert. 127 una: im(m)o, a favourite word with Matthew. 132 rubor: cruor. 133 f. timore, tuwene: timorem, tuuent. 138 uultus: mentis, rather than mu(l)tus. 140 curat (!): twuat, with MS. 141 labellum: labello. 142 uultibus: fletibus. 152 numina: lumina. 163 ansa: ausa (see SPECULUM, 111 (1928), 372). 172 com- paciente loco = 1.34: perhaps flebilibus compatiendo malis (or modis, i.e. elegiac verse). In lines 37, 68, 77, 99 (uitri), 102, 117, 120, 134, 142, 152 Strecker’s conjecture agrees with mine.
II a. Dietrich’s Pyramus, pp. 36 ff. (E. Faral, Sources Lat. des Contes... . du Moyen Age (Paris: Champion, 1913), pp. 41 ff., who often offers the better text), e.g. Il. 6, 9... 191, 208, 225, 248, 250, 253, 266, 283, 291, etc. 1.33; read ut stultos ego subticeam (‘not to mention’) sunt qui sapientes. 51 annos per pueriles, K. S. 53
302 The Textual Criticism of Mediaeval Latin Poets
unmetrical: read gui dum transcendunt annos ambo pueriles. 75 Faral right. 81 unmetrical: read quod et unica? 86 meditansque: meditans quae (i.e. meditans quae uerbula reddat). 96 read te sequor assensum praebens spacit breuioris. 114 Leyser’s conjecture gives the Virgilian but unlikely qui amore: the MS. reading will do, with Faral’s tortantur (hii quod or quot? K.S.) 124 tur is perhaps the end of an obliterated sanguis: other MSS have stopgaps. Faral’s explana- tion will not hold water (see Ovid, Met. iv, 51 ff., and the other ver- sions). 127 udtiua (!): nociua with MS. H. 147 fauit discedere uisus, K. S. 159 fingerat: fingeret. 167 Faral right. 174, the elision is suspicious: perhaps huius ego mortis. 197 genitum, K.S. 191, ete.: see above. 220 mori: mortem? 2320 perhaps nec sinis hunc utta, me neque morte frut: si non uis uita, me sine morte f., K.S. 235 sicut spi- ramina K. S. 263 morte: obviously more. 272 peremptd impossible: read pereunte with one MS. (omit -que). 273 MS. H right (nece crudeli). 277 omit -que. 278 quod monstrans: perhaps quae demens. 285 rapit ... frangit: rapuit . . . fregit. 286 subire MS. H right, but genus seems wrong, as does factus in 288. 300 add comma after ultima. 201 read penultima, with one MS. The next 9 lines are hopeless, and probably spurious. In lines 86, 159, 272, 273, 285 Strecker’s reading coincides with mine; in 167, 191, 208, 220, 225, 246, 248, 250, 269, 286 with Faral’s.
IIb. Anon. Pyramus, Lehmann, p. 46 ff.; Faral, p. 51 ff., 1. 14; add et before opes. 18 laxari, K.S. 40 premit infantum: invert. 91 has no rhyme; perhaps it is two half lines: Il. 91, 92 are unintelligible. 102 MS. H is probably right. 107 prauis seems corrupt. 129 del. comma after nunc. 169 modo fieri: invert.
II c. Anon. Pyramus. (Lehmann offers a shocking text, pp. 52 ff.) 1. 7 sine te nulla potest: nulla potest sine te. 8 edifica moribus: invert. 9 ni: nist (so MSS). 28 pécius (!): parcius (so MSS). 31, insert tacendo after loquendo. 42 read dicit, Nasoque uerus erat. 61 neque eam uagem: perhaps neque uagans abeat (or erret). 62 apta: apte. 76 put semicolon before Aiis nam, not after. 77 nimis: minus. 86 tales: cantus is perhaps preferable to talos. 87 certa: certe. 91 hoc: haec. 92 mouebat: fecit. 102 ignis: ignem. 103 color: calor. 106 -que: atque. 116 studiosa: studiosi. 118 -que: neque. 119 nec: hec. 120
The Textual Criticism of Mediaeval Latin Poets 303
suppresserit: sumpserit. 128 read rigor ille (patrum). 133 read paries after fieret. 146 ad quod: ad os. 152 rumpe: rumpere. 159 perhaps we should read non tamen indignamur: adest uerbum: uia uerbis. 168 exiguo: ex aequo, cf. Ov., Met. iv. 62. 170 respondit inquens: re- spondens inqutt (so A). 179 rimam: rimae. 184 quae: quod. 191 cliuis fons: possibly elutus fons, from Hor., Sat. ii, 4.16 misunderstood, or illimis f. 193 rura: iunci? 195 est umbra: sc(ilicet) umbra. 201: the right sense is got by putting a semicolon after discedunt. 208 um- brosam: ambrosiam (cf. Ov. Met. iv. 215). 209 que: quid. 211 uero: uera. 213 qua: quae. 214 inueniat: ni uenrat. 216 Phebi tolle: invert. 232 leuiter ille minas: illi minas leuiter. 245 Babilonis: Babiloni. 264 put comma before uwmbram, not after. 268 fecit: fecerit. 281 rapido: rabido. 283 ut: et. 284 fuit: fuerit. 296 cernit; discernit (?). 306 f., ‘Your life was worthy to live long (dudum, as Dietrich, |. 231), mine to perish.’ Cf. Ov., Met. iv, 109 f. 313 effundit: effudit. 314 adeste: astate? 321 multo: multis. 326 contracta: cunctata. 344 satis: fontis. 352 add Pyrami before corpus. 356 et . . . morus: ac... mare. 371 actuum: attamen. 379 quae: mthi. 397 manus: manes. 399 reor, sum: sum rea: ero: tbo. 404 frenis: effrenis. 412 sit: sis. 424 superos: pariter?? 425 eddem (as in Propertius, etc.): perhaps tandem. 430 add premit before intus. 432 perhaps emer- gunt curae, answering to discedit somnus. 435 Venus dat: invert. 437 comma after tela, not est. 441 milites: militet. 448 potentis: po- tenter ?
III. De Somnio, Lehmann, pp. 63 ff., 1. 4, que moueant: quod mo- neant. 9 uulnera: fulgida? 19 siccatus: fucatus is here so urgently demanded by the sense, that I would read it, assuming a false quantity. 28 read sordibus et squdlidum pectus hebebat iners? 32 ferus: fétens? 35, MS. H is right. 38 intus: inter. 39 male: mihi. 46 hic . . . comperta: haec composita (unless comperta be neut. plur.). 47 read quod putds ista monent, quae cernis uisa, futura? For quod putds, cf. Thurot, Notices et Extraits, xxi, ii (1868), 400 ff., Vulg., Matt., 24, 45, etc. 53 adsurgere: adspergere. 54 diuitiis . . . tenent: diuitias .. . tenet. 60 hunc: hanc. 67 parvere: parvore.
IVa. Paris Helenae. (pp. 65 ff.) 5 inulta: tnuicta. 10 disparet: despicit?? 21 Tirio: Frigio. 36 prolongant: prérogant? 55 nisi:
304 The Textual Criticism of Mediaeval Latin Poets
quam? unguen: unguem? 89 readand punctuate quid sit; ne deitas tibi constringatur, agendum (sc. est). 100 dit condita: diis c. (‘of heavenly descent’). 135 there should only be a comma at the end. 147 damneque: nam neque. 149 I do not understand. 151 transfer comma to iugales. 204 scutatos: scrutantes is perhaps better than scrutatos. 212 numeretur (nomiretur, MS.): non miretur. 216 autum- nis ut: autumno si? 223 alternantes: altercantes. 224 del. comma.
Helena Paridi. 1. 24 fuitne, rogo: possibly rogo, fuerit. 27 maritat: molestet. 38 nam: namque. 42 quam dicit: quae vicit? 60 praecinis; praecinit. 91 desererem: deserere. 117 lacks a syllable, e.g. (plus) sollicent. 181 cinerata: temerata. 339 lentetur: laetatur. 364 mentem dit: invert.
IVd. Florus Ouidio, etc. (pp. 81 ff.) 1. 38 nulla: perhaps nulli. p- 84 1.7 omit one alter. |. 38 te: me. 47 monimenta recensa (no comma), ‘narrated memories.’ 62 Where does this idea come from?
V. pp. 88f., 1. 3 read hec factt ut placeant que plures querere certant (or placeat quod) (hoc . . . certant, K.S.). 4 quid ertt quod quis amare uelit, K. S.; purpura pallet, Heinze. 5 semicolon at end: no comma after wir in 1. 6. 8 read hec placet: e reliquis nulla suis meritis. 12 que
decorauit, ea: read queque decora, dea, or que decoraret, ea. 13 not clear: perhaps celitus aduentt: media testudine (‘under the middle of the dome’) sedit alta (-e?) sub templt. 16 lentam: letam. 19 sicut: sic tum. 27 reuerentes: reuertentes. 32 before 31; read diua for diuina. 35 uenisti: uenistis. 36 comedere: comesse. $7 dedit: tradidit. 39 f., perhaps (puellis) quas dotem fecit.
Oxford Book of Medieval Latin Verse
Since the above was written a book has apppeared which it is to be hoped will help to raise the standard of mediaeval editing: it is not without trepidation that I venture to make a few suggestions on Mr Stephen Gaselee’s work.
1, 16 nuntiatum: nuntiatur, ‘a rival of his power is announced.’ 20, 122 crebrat: cribrat. 20, 127 natantibys: manantibus? 20, 131 usquam: possibly usque. 20, 232 add semicolon, 234 del. semicolon (circuli depends on ambagibus). 47, 41 et dederunt: exuerunt? (Cant. 5.7). 49, 260 sua: tua? 52, 293 tutae: mutae? 56: stanza vi should
The Textual Criticism of Mediaeval Latin Poets 305
come before v. 57.45 del. full stop (Is., 9.6). 60, 14 super gressus: supergressus (del. comma) ‘transcending physical sense.’ 65.1 clara: data. 65.26 inhiat: nutrical ? 65, 97 et is suspicious, as is the change of persons: read te, and in 96 Aenea. 65, 99 spoponderit: spoponderat. 65, 107 quam: quae. 65, 109 donatae: donata. 69, 18 suppeditas = caleas (from pes)? 71, 20 ornat: fuerat ( = erat), or wernat? 77, 53 patent: parent ( = apparent)? 85.9 rea: sed res? 86, 4 uia: uiae? 88, 19 haec: huic? (reference to the immaculate conception of the Vir- gin). 88.43 plenum: plenam (from the Aue Maria), 89, 13 nec: nune. 102.13 punctuate comedis; si non, (cf. ergo 1. 14). 105.24 baculum: baiulum ? (cf. 1. 39). 108, 15 and 18 tudices tu despicis . . . subiicis.
Wyaceston Boys’ ScHoo., LeIcEsTER, ENGLAND
THE GREGORIAN ANTIPHONARY OF SILOS AND THE SPANISH MELODY OF THE LAMENTATIONS!
R. P. CASIANO ROJO, O.S.B.
HE practical difficulty brought about in Spain in the eleventh
and twelfth centuries by the substitution of the Roman for the Mozarabic liturgy was great and embarrassing; for, as it was neces- sary to shelve the manuscripts of the Spanish rite, the churches found themselves without the indispensable instruments for celebrat- ing the various sacraments and offices of the new use. It is clear that the change was not brought about in a day, nor made simultaneously in all parts of Spain; for it could be accomplished only by the ac- quisition of new books, and this was difficult because of the com- parative scarcity of manuscripts and the time and expense involved in the transcription of copies. Some churches acquired foreign texts; others had copies made with a rapidity remarkable for that time. In general these copies reproduced the script and musical notation of the original manuscript, but there were a few exceptions to the rule. The monks of Silos, although they adopted the Romano- Monastic rite, continued to use Visigothic script and notation for a time, as is proved by two eleventh-century Breviaries, now preserved in the British Museum, in which the whole of the Divine Office ac- cording to the new use is written in the old manner.’ In certain Cata- lan manuscripts, among them an eleventh-century antiphonary now preserved in the ex-Colegiata of San Feliu in Gerona, although French minuscule is used for the text, the musical notation still conserves certain Visigothic elements.* All the other manuscripts
1 Translated from Spanish by Walter Muir Whitehill, Jr. 2 B. M. Add. 30848, Breuiarium de toto (anni) circulo and Add. 30850, Breuiarium seu
Antiphonale Silense. Add. 30850 contains the office of Santo Domingo de Silos, added in an
appendix a few years after his death in 1073. 3 Cf. reproduction in Miguel Rué y Rubié, Canto Gregoriano. Cooperacién & la Edicién Vaticana de Libros de Canto Liturgico (Gerona, 1905), pp. 70, 71.
306
The Gregorian Antiphonary of Silos 307
of the period that I have seen are written in French minuscule with the Aquitanian notation that originated in the south of France and spread to Spain, where it soon took the place of the older Visigothic notation.
Ordinarily the new copies reproduced the text of the original, with the omission of feasts not celebrated in Spain, but without the ad- dition of proper offices for Spanish saints. The reason is obvious: it was not an auspicious period for the composition of new offices and the new Roman rite was not easily accepted by the people. For Spanish feasts the Commune Sanctorum was used, with lessons read from the Passiones that had been used up to that time. In the ad- ministration of the sacraments and in other rites the Mozarabic tradition persisted with more or less stubbornness, as is witnessed by the presence of various older prayers and ceremonies in the manu- scripts of that period.’
In the Liber Ordinum from San Millan de la Cogolla * the Visi- gothic notation of nearly the whole of the Agenda mortuorum, the burial office, has been erased and Aquitaniam neums substituted. There are two possible explanations for the change: either that a Cluniac cantor, obliged to use the manuscript when the Mozarabic liturgy was still current, wished to read the music in a notation bet- ter known to him; or that the Mozarabic office for the burial of the dead was used for some time after the general suppression of the liturgy, so that it was necessary to replace the Visigothic neums by the then general Aquitanian notation. The first hypothesis seems hardly probable; for a similar substitution is found neither in the other pages of the Liber Ordinum nor in any of the other manuscripts from San Millan. The second seems more likely, for the Roman musical formula ‘saeculorum. Amen’ has been added at the end of the antiphons, which would be inexplicable had the change been made before the suppression of the Mozarabic liturgy.
1 Even today in Toledo the burial office differs considerably from the Roman use. ? Madrid, Biblioteca de la Real Academia de la Historia, MS. 56.
The Gregorian Antiphonary of Silos
CONTENTS AND ORIGIN OF THE S1Los MANUSCRIPT
The Silos Antiphonary ' is a manuscript of 396 leaves, measuring 268 x 197 mm., written in double columns of about 34 lines each averaging 175 x 50 mm. It is composed of fifty gatherings of eight leaves each, with the exception of the forty fourth which is of four leaves.
The scheme of the gatherings is as follows: 1 ° (i-viii); 1 * (ix-xvi); ui ® (xvii-xxiv); tv ® (xxv-xxxii); v ® (xxxiii-xl); vi ® (xli-xlviii); vi ° (xlix-lvi); vi ® (Ivii-lxiv); rx * (Ixv-lxxii); x* (Ixxiii-lxxx); x1* (Ixxxi-Ilxxxviii); xm * (Ixxxix-xevi); xm* (xevii-civ); xiv* (eyv- cxii); xv ® (exiii-exx); xvi * (exxi-cxxviii); xv ® (exxix cxxx cxxxi [exxxii] | [exxxiii] exxxiv cxxxv cxxxvi); xvii ® (exxxvii-cexliv); xrx * (i-viii); xx ® (ix-xvi); xx1® (xvii-xxiv); xxm*® (xxv-xxxii); xx! (xxxiii-xl); xxrv ® (xli-xlviii); xxv * (xlix-lvi); [xxv ® (Ivii-lxiv)]; xxvir® (Ixv-lxxii); xxvu1® (Ixxiii-lxxx); xxrx* (Ixxxi-lxxxviii); xxx ® (Ixxxix-xevi); xxx1* (xevii-civ); xxx *® (ev-exii); xxx ° (exiii-cxx); Xxxiv ® (exxi-cxxviii); xxxv* (cxxix-cxxxvi); xxxv1' (exxxvii-cxliv); xxxvi1® (exlv-clii); xxxvu1® (cliii-clx); xxxrx! (elxi-elxviii) ; xL* (elxix-elxxvi) ; xL1° (elxxvii-elxxxiv) ; xLi1* (clxxxv- excii); xLui*® (exciii-cc); xLIv*® (eci-cevili); xLv* (ecix-ccexvi); XLv1‘* (217-220); xuivir® (221-228); xuvimr* (229-236); xix’ (237-244); L® (245-252); 1 ® (253-260); Lx? (261-262, 263 bound in on stubs).
The manuscript appears to have been bound originally in two volumes, as there is an accurate foliation in red Roman numerals, contemporary with the manuscript and in two series. Gatherings I — xvii are numbered from i to cxliv; gatherings x1x — xLv are in a separate series from i to ccxvi, while gatherings xLvI — LII are unnumbered. A few leaves are wanting at the end.?
The manuscript contains the following books —in an earlier period written separately — necessary for the recitation of a com- plete office: the Antiphonarium, composed of the antiphons and re- sponses; the Lectionarium, with the selections from the Bible and
1 MS. 9 in the Archives of Santo Domingo de Silos.
2 There is no modern foliation, and accordingly, to distinguish the two series of numbers, those of the second foliation will be referred to in the following manner: fol. rriv°.
The Gregorian Antiphonary of Silos 309
other books that are read as lessons; the Psalterium containing the necessary Psalms and Canticles; and, at the end, the Hymnarium. The months of March and April of the Proprium Sanctorum ' and the end of the Hymnarium are wanting.
There is no colophon or subscription that gives the precise date and place of composition, but the script seems to be of the end of the twelfth or the beginning of the thirteenth century.2 The manu- script is written in a superbly regular French minuscule that gives almost the effect of a printed page: it seems likely that it was copied directly from a French original, in so much as the punctuation and the abbreviations are entirely French and show no influence of Visigothic usage. This is confirmed by the presence in the text of a number of homilies ascribed to Dom Herico or Heyricus, a ninth century monk of St Germain d’ Auxerre,* and the offices of French saints such as St Radegonde and St Brice, Bishop of Tours.
The manuscript is of Spanish origin, or at least written for use in Spain, as may be seen from an examination of the Calendar, which contains two feasts for St James the Great, that of July 25, cele- brated with an octave, and the Translation, observed on December 30. The Proprium Sanctorum contains a great number of Spanish saints, many of whom were not honored in other countries, such as SS. Fructuoso, Augurio, and Eulogio, St Ildefonso of Toledo, St Eulalia de Barcelona, St Leandro, St Rosendo, St Prudencio, St Torcuato and companions, St Pelagio, St Zoilo, St Marina, St Justa and St Rufina, St Cucufato, SS. Verisimo, Maxima, and Julia, SS. Vincente, Sabina, and Cristeta, SS. Lupercio, Claudio, and Victorino, St Millan, SS. Acisclo and Victoria, SS. Facundo and Primitivo, St Leocadia, St Eulalia de Mérida, the Commemoratio Sanctae Mariae of December 18, and the translation of St Isidore. The sung portions of these offices are, in general, taken from the Common of Saints,
1 Contained in the missing gathering, foll. lvii°-Ixiv’.
* The manuscript cannot be anterior to 1170, as it contains a commemoration of the translation of S. Rosendo, which took place in that year. Cf. Antonio de Yepes, Coronica General de la Orden de San Benito (Valladolid, 1615), fol. 14.
* This Heyricus wrote some 64 homilies, the introduction to which is published in the Patrologia Latina, cxx1v, 1129-1182. He was later honored as a saint, but in the manu- script he is referred to only as Dom.
310 The Gregorian Antiphonary of Silos
but the manuscript contains two new antiphons in honor of St James in which he is invoked with the title of patrone singularis, The lessons and homilies are taken from the Passiones that were read in Spanish churches during the Mozarabic period."
The offices are arranged according to the monastic use, but there is no indication that the manuscript was originally intended for use at Silos. It is true that an office is provided for St Sebastian, patron of the Silos church, but it is joined to the commemoration of St Fabian, as is the universal custom: there is no mention of Santo Domingo de Silos, the patron of the monastery, for whom an office is always provided in Silos manuscripts.? On the contrary, there are strong arguments for the monastery of San Salvador de Celanova in Galicia. The first of these is that the manuscript contains a proper office for the feast of the Transfiguration, the patronal feast of Celanova. This observance, which is of eastern origin, did not be- come general in the western church until 1456, and, although it was celebrated from the twelfth century on in monasteries dependent on Cluny, from which it might have spread to an independent house like Celanova, it seems that the importance given to it in this manu- script, where an octave is provided with a great number of pices for the days infra octauam, must be the result of a local observance. The second argument in favour of Celanova is the presence not only of the feast of St Rosendo, the founder and patron of the monastery, but also an office commemorating his translation, an observance clearly of local origin which, according to Yepes * was celebrated in Celanova. Finally this attribution explains the presence of an of- fice for St Giraldo, a native of southern France, at one time Precentor of Toledo, who later became Archbishop of Braga, the ecclesiastical province to which Celanova belonged.
Musically the manuscript is of considerable importance. It was one of the principal sources consulted by Dom Pothier in preparing
1 The authors of some of these lessons are known; that of St Ildefonso was composed by Cixila, an eighth-century Archbishop of Toledo, and that of St Pelayo is the work of a Galician priest, Raguel.
2 Cf. Brit. Mus., Add. 30850, foll. 219-221, which is of the eleventh century.
3 Antonio de Yepes, Coronica General de la Orden de San Benito, v, fol. 15.
The Gregorian Antiphonary of Silos 311
the 1891 edition of the Solesmes Antiphonary,' that has more recently become the basis of the present official Vatican edition. The nota- tion is of the Aquitanian superposed point system, written with a single red line, but with the diastematic intervals so carefully cal- culated that it can be read with the same rapidity as a modern musical text with a complete staff. The Gregorian tradition repre- sented is good and lawful, although like all the manuscripts of this period there is a certain tendency toward the embellishment and elaboration of the melody and the notes.
Among the music which it contains are several unpublished pieces, worthy of being better known; not the least interesting of these are two antiphons in honor of the patron of Spain, St James, which probably date from the suppression of the Mozarabic liturgy. It is unfortunate that they have not been included in the office of St James that is at present in use.
IN NATALE S. IACOBI I. VESP. ap MAGNIFICAT. (1).
_.
&- te Ia- cé6- be, ye) ‘a
~ -
um cor - de,
ad Dé- mi- num.
1 Liber Antiphonarius pro diurnis horis juxta ritum monasticum kalendario Ordinis Sancti Benedicti accomodatus (Solesmes: Imprimerie Saint-Pierre, 1891).
The Gregorian Antiphonary of Silos II. VESP. ap MAGNIFICAT (2).
1 — = a4, | ro rAd hc I °
San- cte Ia- cé - be , inter- cé-
a | i TR Mk 7
a de pro no-bis, ut con-sor- tes gl6- ri- ae
§ ; Ap | a & -~ a @ Pee * oi man - San - ct6- rum te - ef- fi- ci me-
€-
a. a & re- A- mur. (1) ef. fol. XCVIv.* (2) fol. XCVIIv.*
Tue LAMENTATIONS
The manuscript contains four special tones for the Lamentations of Holy Week and one for the Oratio Ieremiae, which I have already published with three of the tones of the Lamentations; ' the fourth is unpublished, and appears for the first time at the end of this article. For convenience of reference the beginning of the three other tones and the Oratio are also given at the end.
Various Spanish manuscripts anterior and posterior to the Silos Antiphonary contain similar melodies of the Lamentations. Only the first and simplest formula — from which the other three and many other melodies of unequal value found in Spanish manuscripts developed — is found in the earliest manuscripts. The Visigothic Bible of San Pedro de Cardefia,? dating from the tenth or eleventh century, which is now preserved in the Universidad Pontificia in
1 Cantus Lamentationum apud Hispanos usurpatus quem ex codice Silensi saeculo XIII conscripto nunc primum iuris publice fecit (Bilbao, 1917). Copies may still be had from “Orfeo
Tracio” (n. d.), Génova, 19, Madrid. 2 Cf. A. Andrés, Bolettn de la Real Academia de la Historia tx (1912), 101-146.
The Gregorian Antiphonary of Silos ' 318
Burgos, contains in foll. 236 and 237 with the first two chapters of the Lamentations of Jeremiah a simple melody in Mozarabic neums that, to all appearances, is that of the first melody of the Silos manuscript. This is also found in a Lectionary, now preserved in the collection of D. Narciso Sambola in Gerona, written in the eleventh-century with the superposed point notation.! A thirteenth- century Antiphonary in the Archives of the Cathedral of Huesca contains two tones, still simple although slightly more elaborate than those of the Cardefia and Gerona manuscripts: one, appointed for Maundy Thursday, is similar to No. 1, without the formula of the third phrase; the other, used for the Lamentations of Holy Saturday and for the Oratio Ieremiae, is similar to No. m1, but in both the cadences are shorter and a little monotonous.
In the Archives of the Cathedral of Toledo there is a fourteenth- century Breviary? that contains the tones of the Lamentations. The first (Maundy Thursday) resembles No. 11, but with the order of the first and second cadences interchanged: the final phrase is entirely different, and resembles the melody of the Ezrultet found in many Spanish missals. The first and second tones of Maundy Thursday somewhat resemble No. 1 in the general theme, but are quite different in the greater part of the cadences. The other tones of this manuscript have no resemblance to the Silos manuscript, nor are they found in older documents: their character is that of a much more modern period.
Since the sixteenth century all the Spanish cathedrals and col- legiate churches have possessed beautiful, and sometimes richly decorated, Pasionarios * that contained various tones for the Lam- entations, the greater part of which reproduced more or less faith- fully those of the previously mentioned Toledo manuscript: however, nearly all of these added some new tones, some of which, being little more than old formulas put to new uses, are of slight interest, while others are actually new and of greater importance.
1 Three columns of this manuscript are reproduced by Rué y Rubi, op. cit., pp. 66, 67
2 MS. 35.9.
* So called because, among the other things, the account of the Passion according to the four Gospels occupied the principal place.
314 The Gregorian Antiphonary of Silos
A fifteenth-century Troper, No. 23 of the Cathedral Archives of Leén, written in a heavy superposed point notation with a single line, concludes with a series of Lamentations, the richest in variants of any that I have seen, though many of them bear little relation to the traditional tones and date from the period in which the manu- script was written.
However, none of the other manuscripts containing the Lam- entations mentioned in this study can be compared to the Silos Antiphonary for the combination of truly classical restraint and all the variety possible in this type of music: these melodies possess the unity and the natural flowing quality of the best Gregorian com- positions. The general tone is one of serenity and resignation.
ORIGIN OF THE S1Los LAMENTATIONS
In their general nature and scale these melodies resemble funda- mentally many Roman recitatives, such as the oldest melody of the Collects (ad libitum, No. 1 in the Vatican edition),' and those of the Prefaces and the Pater Noster: the details in which the Lamentations differ from these Roman melodies are all found in the Gregorian antiphonary.
Unfortunately it is impossible to trace a satisfactory relationship with authentic Mozarabic melodies, as so few have been deciphered, but, notwithstanding, in those that are known some of the motives of the Lamentations are found. The first part of the Oratio Ieremiae, Recordare Domine quid acciderit nobis is identical, note for note, with the beginning of the two antiphons Si ascendero in caelum and Manus tua Deus deducet me of the Mozarabic Agenda mortuorum.’
The Oferencio (Ordo Missae) given by Cisneros to the Mozarabic Chapel of Corpus Christi in Toledo Cathedral in the parts relating to music is inspired by the same fundamental themes as the Lam- entations and the oldest recitatives of the Roman rite, and more
1 Graduale Sacrosanctae Romanae Ecclesiae de Tempore et de Sanctis (Rome: Tipografia Vaticana, 1908), pp. 96*-97.*
2 Cf. R. P. Casiano Rojo and R. P. German Prado, El Canto Mozdrabe, Estudio hist6rico- erttico de su Antigiiedad y Estado actual (Barcelona: Diputaci6n Provincial, 1929), also R. P. German Prado, “ Mozarabic Melodics,” Specutum, u1 (1928), 236, 237 and 1v (1929), 145. The theme of Recordare is also found on p. 230 applied to the text Per gloriam nominis tui.
The Gregorian Antiphonary of Silos 315
particularly contains many of the formulas and cadences of the tones that we are studying. The prayer Per gloriam nominis tui that was solemnly sung by the celebrant on feast days at the beginning of the Mozarabic mass follows the same melodic lines as No. 11, but differs from it in the final cadence and includes various formulas of No. m1 and No. tv. However, we are not standing on particularly firm ground with the books of Cisneros, for the Mozarabic origin of all their contents is doubtful. When he undertook his liturgical resto- ration at the beginning of the sixteenth century the Mozarabic tradition was already completely lost. The two great Cantorales of the Chapel of Corpus Christi probably contain few authentic Moz- arabic melodies: the Laudes contain some, but in a sufficiently adulterated form. It is, in consequence, difficult to determine the exact value of the Oferencio, especially as we have no manuscript containing the Common melodies of the Mozarabic rite. It may be that, in spite of the decadence of the rite in the years that preceded the restoration of Cisneros, some memory of these melodies or some document, now lost, was preserved,' by means of which the Oferencio may contain, in substance at least, some of the tones used in the best period of Mozarabic music: on the other hand, it is equally possible that, wanting authentic Mozarabic melodies, Cisneros adapted various chants of the Hispano-Roman rite then in use for his new books. In any event, before his time, we find in the books of this Hispano-Roman rite some of the melodies which he included in his Mozarabic revision, for the tone of the Per gloriam nominis tut is exactly the same as that of the Holy Saturday Praeconium Paschale which is found in Spanish missals and Pasionarios both be- fore and after Cisneros.
Nevertheless, in spite of this deficiency of authentic Mozarabic melodies for comparison, the question of the origin of these tones is capable of solution, for the melodies, at least in part, are found in various non-Spanish manuscripts. In reality these formulas, at least in their fundamental motives, form part of the common prop- erty of all the western liturgies. In other countries, if these tones
! The preservation of the melodies of the Agenda mortuorum in the San Mill4n Liber Ordinum makes such a notion seem possible.
316 The Gregorian Antiphonary of Silos
were applied to the Lamentations at any early date, they were abandoned in favour of that which has recently been adopted in the Vatican edition ' for use in the universal church, which is said to be the same tone used by the Jews in their synagogues. In Spain, to the contrary, they have been used to the almost complete exclusion of the other, and are not only found in the Gregorian manuscripts but are reproduced in modern books as well.
The following illustrations give the beginning and the end of the melodies as found in the Silos manuscript with the defects found in the original. The first defect is that quite without necessity the notes of the melody are frequently doubled in the vocalizations and in the syllabic parts.? The second defect is the use of the liquescent note in circumstances in which it has neither reason nor effect. The copyist of the manuscript seems to have regarded the liquescent as a device for tying together, an ascending note, by means of which in the syllabic chants the voice passes from one note to the following, giv- ing a sort of anticipation of the following note before actually reach- ing it. This is not the function of the liquescent, which is properly employed only with two consecutive consonants or two vowels forming a dipthong, when, to pronounce the second properly, it is necessary to pass to the note of the syllable following: by this means sonority and clear articulation are lost, but the time value is pre- served. The copyist has used the small note that expresses the liquescent in places in which it is unnecessary, as there are neither two consonants nor dipthongs. In No. tv, with the word candidiores the liquescent is well placed; in rubteundiores and denigrata and in the Ierusalem of the other pieces it has no object.
The reader will note that in the word manuum, toward the end of No. 111, the notes are arranged as if it were a word of two syllables only, ma-num. This procedure is correct, and was used in all
1 Officiuen Maioris Hebdomadae et Octauae Paschae a Dominicis in Palmis usque ad Sab-
batum in Albis cum cantu. Editio Typica Vaticana (Rome: Tipografia Poliglotta Vaticana, 1922), pp. 283-289, 366-372, 487-443.
2 Cf. the formula that accompanies the letter Jod in No. 1 and the final words of each strophe: cf. also the words ingressus, praecepetas, Ierusalem. In the other melodies there are unnecessary notes that hinder the presentation of the text.
3 For a more complete explanation of this vocal effect see my Metédo de Canto Gregoriano
(Santo Domingo de Silos, 1906), pp. 66, 67.
The Gregorian Antiphonary of Silos 317
manuscripts. When a vowel is repeated, and the first accompanies a group of notes, the second repeated vowel is not pronounced sep- arately, but is considered to have been dealt with sufficiently in the vocalization of the first.
CANTUS LAMENTATIONUM CANTUS I Feria v. in Caena Domm. (1)
a Qué- mo- do se- det
* @ 5
. ple- na pé- pu- lo:
a, re
ae é a a. = sal -
fa- cta est qua- si wui- du-
a
nti- um: prin-ceps pro- uin- ci-
*_3
est sub tri- bu- to.
aa 4 4 aff LA
me rs] x
sa- lem, JTe- ra- sa- lem, conuér -
> A
a
a iebimeiieaisaniannaal fe
+. =
te- re ad Dé- mi num De- um tu- um.
(1) ef. fol. LXIIIr and v.
CANTUS II Feria vi. in Parasceue. (1).
7 —tHhra ae i 3. a 2 o ? Jod. Ma- num su- am mi-
Z vo ae
in- grés-sas San- ctu -
2 +
= « *
qui- bus prae- cé- pe- ras
. Lal
” ~ hall Lal *
a “. =
in ec- clé- si- am tu- i 7
lem, Ie- ria-
a. r~ = Lad we
ad Démi- num De- um tu-
(1) cf. fol. LXVv.
CANTUS Ill Sabbato Sancto. (1).
Paid t,24, ys .
a) Qué - mo - do
leph
tum est au- rum,
3 , 4 lt « "A, id I eli hel
pti- mus, 3 A.
la- pi- des sanctu -
mi- cti au-ro_ pri- mo: Qué- mo- do
, ha =. bal — hanl
a =
a * -
re- pu- t&a- ti sunt in ua-
£ &
sa té- ste- a,
0 - pus ma- nuum fi- gu-
1 +
sa-lem, TJTe- ri- sa- lem, conuér- te- re ad
Dé- mi- num De- um tu - (1) ef. fol. LXVII.
The Gregorian Antiphonary of Silos
. fa" s — > An a '-¢,8A,
vt
ter- fé- ctis fa-
“1¢,8 - im 4 -_
y ter - rae.
(> af. pp zs a
ra- sa- lem, con-uér -te -
[bh - Ea
Dé6- mi- num De- um tu- um.
AN UNPUBLISHED MELODY OF THE LAMENTATIONS
In the Silos manuscript in the office for Holy Saturday, after the Oratio Ieremiae and with no other title than Alia, is a lesson, the text of which is not found in the modern Breviary, apparently in- tended to be used ad libitum as the third lesson of the first nocturn in place of the Oratio Ieremiae. It is taken from the fourth chapter of the Lamentations, and is composed of the verses that follow those of the second lesson of this nocturn.
ALIUS CANTUS LAMENTATIONIS SABBATO SANCTO. (1).
¢ fen t Seals y t.1. el,
4,20, _
Can-di- di- 6- res
ius ni- ue,
ru- bi-cun-di- 6-res
est fa- ci- es e- 6-rum - su-
car- bé- nes, et non sunt cég- ni-ti
ad- haé- sit
Tu - it,
(1) cf. fol.
322 The Gregorian Anttphonary of Silos
The melody in no way resembles the preceding tones, neither is it similar to anything in the Mozarabic or Gregorian repertories. Its closest relationship seems to be with the following antiphon for Sab- bato in Parasceve in the Ambrosian Antiphonary.? The fact that both pieces are destined for use in Mattins of the same day,'the one as a lesson, the other as an antiphon, is a coincidence worth noting.
SABBATO IN PARASCEVE AD MATUTINUM. ANT. (1)
o P6- su- i ue- sti- mén- tum me- um — a j}—_¢—__9°_» _» _»_» _y ci- Ii- ci- um, et fa- ctus sum il- lis 6 o——_-—-v ose a ss ai a os. a * s e in pa- ra- bo- lam..&. A-d-uér- sum me é
——@-—"_8@"""8"""“8 2.0200 nnn Bs sO OS" a
ex - er- ce- ban- tur qui se- dé- bant ¢—*— st _ ete 7 in por- ta, et in me psal- lé- bant ¢ - - —_ 2 o—_—*# na 2 ae s_ > qui bi- bé- bant.ui- num e- go ue- — 2 Reon os - - = s “ as 4 ro o- ra- ti- 6- nem me- am ad ¢ I
te, Dé-mi- ne.
(1) cf. Paléographie musicale, VI, 302.
Qo 4
= il @ « in a ~-/nn? n ——)
a
The Gregorian Antiphonary of Silos 323
The melody is in the first mode, and is of an expansive character that is in sharp contrast to the other tones in the manuscript: it ap- pears to have been inspired by the first strophe of the text, Candi- diores Nazaraet eius niue, nitidiores lacte, rubicundiores ebore antiquo, saphiro pulchriores. It is probably of the period of the manuscript, although it may possibly be a little older: the use of the first mode scale with the constant introduction of si-natural, and the elaborate vocalizations at the ends of the strophes are characteristic of this period. Although this text is not contained in the present day Breviary, the melody could easily be applied to the second lesson of Holy Saturday, Quomodo obscuratum est aurum, as the strophes of both texts are of similar character.
1 F, lvii®.
2 British Museum, Add. 34209, reproduced in facsimile in Paléographie Musicale (Soles- mes, 1), v, 253, 254. Cf. the text, v1, 302.
Reat ABADiA DE Santo DomINGo DE SILOs, Burcos, Spain
CANTUS IV Sabbato Sancto. (1).
t. 2. § ——— fis —3—fea—,- tL
ee: a _ O- ri- ti- o Ile- re-mfi- ae Prophé- a
a tae. Re- cor- da- re, Dé-mi- ne, quid ac- ci-
od oe . de- rit no- bis: in- tu- é-_ re, et Z L/S ee poops & mn & ry.¥ io ré- spi- ce op- pré- bri- um_ no- strum. RRS: NE a 2 a ae r | vo id He- ré- di- tas no- stra uer- sa est rs - & s. 4 eo . * “t +t A. | + Q ‘ ee * 7 , ad a- li- é- nos: do- mus no- strae , I rs 7 ee v Flin a ee {® - v ] ap * aie ] ™ ‘ ad ex - tra- ne- os. Te- rii- sa-_ lem, eg o a 4 a a ' —— a a a Ae oe : 7 ¥ — +, é Ie- ri-sa- lem, con-uér- te- re ad c a t a A e' € “ad | ~~ = e
“
Dé6- mi- num De- um tu- um.
(1) cf. fol. LVIIr and v.
7 I N
REVIEWS
Anniversary Essays in Mediaeval History by Students of Charles Homer Haskins presented on His Completion of Forty Years of Teaching. Boston: Houghton Mifflin Co. Cloth. Pp.x + 417. $5.00.
Amonc American Festschriften which have appeared, this is a notable vol- ume distinguished both by the eminence of the scholar to whom it is ded- icated and the quality of the contributions. If not by reason of years, certainly by distinction of scholarship, Professor Haskins is the doyen of American mediaevalists, doubly remarkable as a creative and productive scholar and as a maker of students and teachers to follow in his footsteps. There are eighteen contributions to this volume, one of them by a former Radcliffe College student. An analysis of the nature of these eighteen sub- jects is interesting, for thereby one sees how the dominant interests of Pro- fessor Haskins as an historian are reflected in the work of his students. The intellectual history of mediaeval Europe and mediaeval institutional history are almost evenly divided, there being six of the former and seven of the latter. When one remembers the Norman Institutions and The Renaissance of the Twelfth Century, one perceives how true to the master’s ideals his students are. It is odd, though, not to find any study on the history of mediaeval science in this list. The rest of the subjects comprise three of a political nature and two on law. As to geographical distribution the authors are interestingly distributed: Massachusetts five, Pennsylvania three, New York three, and New Jersey, New Hampshire, Wisconsin, Illinois, Iowa, Nebraska, and New Mexico each one. Evidently the New England and Middle Atlantic states may be regarded as Professor Haskins’ domain. But as a scholar every mediaevalist knows that his line is drawn out to the ends of the earth and his words to the end of the world. Europe claims him as do we of America.
Not every one of these essays is as deep as a well or as wide as a church door. But as a whole the quality of them all is very high. Some of the con- tributors are more seasoned scholars than others, so that one naturally dis- covers some degree of unevennness among the essays. If the reviewer may express a preference, he would especially designate Mr Howard L. Gray’s ‘Greek Visitors to England in 1455-56’; Mr William E. Lunt’s ‘Clerical Tenths Levied in England by Papal Authority during the Reign of Edward II’; Mr Sidney R. Packard’s ‘Norman Communes under Richard and John’; Mr Carl Stephenson’s ‘Taxation and Representation in the Middle Ages’;
325
326 Reviews
and Mr C. H. Taylor’s ‘Census de rebus in the Capitularies.’ With one ex- ception, this selection puts the contributions pertaining to institutional history upon the preferred list. But no invidious distinction is implied. It may be that the intellectual history of the Middle Ages is harder to under- stand than its institutional history; yet it would seem also that the writers of the essays which have to deal with the former subject are less experienced scholars than some of the other contributors.
The volume has been edited by Mr C. H. Taylor, Assistant Professor of history in Harvard. A portrait of Professor Haskins which every possessor of the volume will cherish is in the front of the volume, which is terminated with a bibliography of Professor Haskins’ writings which fortunately in- cludes even his book reviews.
JAMES WESTFALL THOMPSON, University of Chicago
1. Branv and N. Carrosan, Album de Paleografie Romaneasca (Scrierea Chirilica). Bucharest: Socec and Co., 1929. Boards. 35 plates. Lei 170.
Ir has been observed that the use of the Cyrillic alphabet among the Ru- manians is one of the most interesting moments in the history of Graeco- Slavic civilization. The adoption of this alphabet is initially traceable to Bulgarian influence. Even in the fourteenth century, after the foundation of the independent principalities of Moldavia and Wallachia, the Rumanians continued to maintain their intimate contact with Slavic culture, the dom- inance of which did not begin to decline until some 150 years later. It is thus in the early sixteenth century that Rumanian literary monuments be- gin to appear, composed in the vernacular, but written in the Cyrillic script. The habit of introducing formal documents with a Church Slavic formula persisted, as a matter of fact, in Moldavia as late as the second half of the eighteenth century, and the number of Rumanian texts written or printed in the Latin alphabet prior to 1780 is extremely slight. It is also undeniable that, from the phonetic standpoint, the Cyrillic alphabet was better adapted to the representation of Rumanian sounds than the Latin. It thus favored the evolution of an exact and constant orthography. As Tiktin has re- marked (Groéber’s Grundriss, I (2d ed.), 571):
Der rein phonetische Charakter der cyrillisch-ruminischen Schrift begiinstigte die allmihliche Ausbildung einer auf klaren und festen wenn auch in keiner Gram- matik kodificierten Regeln beruhenden Orthographie, welche etwa um den Beginn des vorigen [18ten] Jahrhunderts sich so allgemeine Geltung verschafft hatte, dass
die Urkunden dieses und der ersten Hiilfte des gegenwirtigen [19ten] Jahrhunderts insgemein kaum nennenswerte orthographische Divergenzen aufweisen.
The attachment of the Rumanians to the Cyrillic alphabet was indeed so great that despite the multiplication of Greek schools and influence under
Reviews 327
the Phanariote régime, they continued to use this alphabet not only in vernacular documents but also for instruments composed in Greek. The initial impulse toward the adoption of the Latin alphabet proceeded from the Uniate clergy in Siebenbiirgen, in connection with their Catholic prop- aganda and a related desire to emphasize the Romanic character of the Rumanian nation. A natural resistance by the Orthodox elements of the population delayed the realization of this reform, so that only in 1860 did the Latin alphabet become the sole script for school and official use. It was adopted considerably later by the ecclesiastical authorities, and even shortly before the World War, calendars and various works of popular nature were still being printed in the Tsirillitsa, which was especially prev- alent in Bessarabia.
In the light of these circumstances, it is clear that Cyrillic palaeography is an important auxiliary to the study of Rumanian philology. The Album of Professors Bianu and Cartojan (though the specimens in general offer little novelty to the student of Slavic palaeography) provides a convenient instrument for this branch of investigation. Beginning with the Psaltirea Scheiana and the Voronetul Codex, the compilers offer a representative series of facsimiles extending from semi-uncials of the sixteenth century to cursives of the nineteenth, together with several interesting plates of liga- tures, initials, and tail-pieces. It would perhaps have been worth while to include one or two specimens of contemporary South-Slavic Cyrillic texts to parallel the script of the Psaltirea Scheiana. One might profitably com- pare the Prophets written at Athos in 1542-1543 (cf. Lavrov: [aseo- epaguueckoe O6ospmnie, no. 303, p. 275), now in the National Library at Belgrade (No. 26), or the Lomnica Tipik of 1578 (Lavrov, op. cit., no. 305, p. 277; Belgrade Academy of Sciences, No. 120). In any case, the present collection will doubtless form a valuable adjunct to seminary studies in this comparatively neglected subdivision of Romance Philology, which will be still further promoted by the publication of the manual of Rumanian Cyril- lic palaeography which these eminent scholars have in preparation.
SamuEL H. Cross, Harvard University
F. HottHausen, ed., Beowulf nebst den kleineren Denkmdlern der Heldensage. Heidelberg: Carl Winter, 1929. I. Teil: Texte und Namenverzeichnis, 6. verbesserte Auflage, pp. x + 123; II. Teil: Einleitung, Glossar und Anmerkungen, 5. verbesserte Auflage, pp. xl+ 214. Paper. Mk. 5.
Tue texts which Holthausen prints in the present edition are Beowulf, the Finnsburg fragment, the two Waldere fragments, Deor and Widsith. Pre- fixed are four plates, which give us an idea of the handwriting of as many
328 Reviews
scribes (including, of course, the two hands of Beowulf). Texts and Namen- verzeichnis alike seem to be a Manuldruck based on the plates of the edition of 1921, though there is no mention of the point in the Vorwort. It must be added that a few changes were apparently made on the plates, so that the sixth edition is not identical with the fifth. As regards Part II, the Finleit- ung is reprinted from the edition of 1919, but seven pages of Nachtrdge are annexed. The glossary and the index are described as reprints ‘mit Plat- tenkorrekturen.’ The notes, however, were ‘vollstdndig neu gesetzt.’
The editor tells us (I, vi) that his texts differ from those of the previous edition in two ways. On the one hand, he has restored defensible manu- script readings; on the other, he has tried to correct, by emendation, some corrupt passages. The result of his editorial work cannot be described as an edition that holds fast to the iiberlieferung. I have gone through the first 500 lines of Beowulf as Holthausen prints it, and find no less than 98 in- stances of what I consider needless emendation, to which other students of the poem might well add a few further instances. And yet it would be un- just to say that Holthausen is an erratic or irresponsible editor of Old-Eng- lish texts. He may take more liberties with the manuscripts than do some others, but he belongs to the authentic tradition of Old-English philology, which has never shown much reverence for the monuments as they are. The so-called conservatism of certain Old-English philologists exhibits it- self not so much in a reluctance to emend as in a reluctance to give up old emendations in favor of the manuscript reading! Holthausen’s edition is full of this kind of ‘conservatism.’ Indeed, most of the emendations that true conservatives, like myself, object to are old conjectures, handed down piously from generation to generation and long since become so sacred that any mention of the actual reading of the manuscript shocks and alarms the faithful. I note that Holthausen keeps his punctum delens under the 8 of dod in |. 1231, and I am not surprised, for fidelity to the text, in the face of earlier emendation, is not yet reckoned a virtue in Old-English scholarship.
Kemp MALOong, Johns Hopkins University
Ca. Petit Dutartuis and Grorces Leresvre, Studies and Notes Supplementary to Stubbs’ Constitutional History, translated by M. I. E. Robertson and R. F. Treharne. Manchester: University Press, 1929. Cloth. Pp. ix, 303-517.
ALTHOUGH these studies are associated in publication with the third volume of Stubbs, they relate almost entirely to subjects covered in his first two volumes. In the introductory chapter, M. Petit Dutaillis attempts to dis- abuse the reader of the belief held by Stubbs, that ‘Edward I . . . definitely founded the parliamentary monarchy’ (p. 308). It is a drastic criticism.
. -™ eS 65 6 me ele
oe s&s 2. ®
sa 3S? sa oh s™| & © ff oO VM +
Reviews 329
Indeed, some of his arguments are more emphatic than his conclusion de- mands. Scholars who would agree that ‘the Parliament of 1305 . . . owes its new features to the perfecting of monarchical centralisation [sic]’ (p. 346) — possibly reserving the privilege of slight modification — might not accept as proved either that, during the whole of the Middle Ages, the com- mons contributed nothing to the foundation of the constitutional monarchy (p. $25), or that all the English kings of the last centuries of the Middle Ages sought to follow a policy of megalomania (p. 313).
M. Lefebvre traces the development of the administration and the origins of the House of Commons as they are depicted in the literature published approximately before 1928. He correlates and interprets the new evidence and opinions, reducing them to a brief summary of remarkable clarity. He also illumines the subject with analogies and contrasts drawn from French institutions. His treatment results in a series of judgments with which the reader of the literature on which they are based is likely to find himself sometimes in agreement and sometimes not. Beyond such matters of opinion there is a modicum of inaccuracies, such as are nearly inevitable in a brief generalization of so large a mass of complicated materials. The chancery, for example, hardly became a court of common law (pp. 380, 406). Richardson and Sayles, in the Bulletin of the Institute of Historical Research (v1, 76) demonstrate the error of the statement that ‘under Ed- ward II men began to reserve the name “Council in Parliament” for the assemblies in which the barons summoned to the Council sat conjointly with the Commons’ (p. 434), if there was doubt previously. Surely it is too categorical to assert, without qualification and without adequate consid- eration of interpretations to the contrary, that Edward I ‘did not think of conceding to the Commons the right of consenting to taxation’ (p. 484).
There are several typographical errors and apparently one mistake in translation. On page 332, line 13, the sense seems to require the substitu- tion of ‘any’ for ‘no.’
For one who is beginning the study of the English Constitution this book provides a valuable outline of recent discoveries and a guide to the literature. Even seasoned students may find it useful. In a field where the literature is increasing and will increase so rapidly, an estimate made by two such competent and impartial scholars of the stage which our knowledge had reached at a definite time can hardly fail to be enlightening.
W. E. Lunt, Haverford College
330 Reviews
Epwarp Kennarp Rano, Studies in the Script of Tours, I; A Survey of the Manuscripts of Tours. Cambridge, Mass.: The Mediaeval Academy of America, Publication No. 3, 1929. Vol. 1, Text. Pp. xxi-+ 245; vol. u, Plates CC. $50.00.
Tuis is not only an important work in the new material and conclusions it brings to the student of palaeography, but it is still more a milestone in the progress of the study of mediaeval manuscripts, and an incentive to other scholars to carry on and enrich the study of this and allied fields. The pioneer work of Delisle and Traube, together with Rand’s earlier studies and the contributions of Kéhler and others, are all utilized to make this picture of the manuscripts of Tours as complete and accurate as possible. I remember that Traube, in his lectures in 1896, pictured works like this as the next great task for Latin palaeographers. He then hoped to see not one but many mediaeval schools of writing studied and illustrated with ade- quate facsimiles. Most of this work is still to be done, but the need at least is now generally recognized.
In Rand’s work the chief weight has been laid on the styles of writing in the eighth and ninth centuries, but the entire activity of these three monas- tic schools from the sixth to the twelfth century has been included.
The first two chapters treat of the libraries and the general characteris- tics of the script. Chapter ITI is an especially important contribution cover- ing the size, ruling, quires, signatures, abbreviations, punctuation, and text of the manuscripts studied. Much of this comprehensive survey is derived from Rand’s earlier studies, and the section on the systems of ruling is of the greatest value for the whole investigation.
Chapter IV gives the stages in the development of the Script of Tours. Only suggestions are given on what style of writing prevailed in the sixth and seventh centuries. The Ashburnham Pentateuch is claimed for the end of this period, whether rightly or not is still a question, but the manuscript was certainly at Tours later. With the eighth century, the study becomes more exact. The different styles of writing are classified under the headings: the Irish at Tours, the Pre-Alcuinian Style, and the Reforms of Alcuin. The last of these is of special interest, for against Traube’s dictum, Rand maintains definite and extensive influence of Alcuin. The manuscripts described as the Embellished Merovingian Style are assigned to the time of Alcuin, or immediately thereafter, while the so-called Regular Style seems to follow Alcuin.
The ninth century shows a far greater number of manuscripts definitely recognized and dated. Six types of writing are distinguished and assigned to their proper succession in the century, though these sub-periods natur- ally overlapped somewhat. Also in this period enough books are preserved so that the three schools of Tours, St Martin, St Gatian, and Marmoutier, can be distinguished in their styles of writing.
Reviews 331
The survey is carried on through the tenth, eleventh, and twelfth centur- ries, but is more general in character as the number of manuscripts thus far recognized and assigned to definite dates and monasteries is limited. Fur- thermore, no such intensive study of the manuscripts listed for these cen- turies has as yet been attempted.
After these historical chapters there follows a full, systematic description of each manuscript studied, a model for workers in palaeography. Very rarely is any essential item omitted or any point treated left without a definite statement, or at least an expression of opinion. When neither is given, there is regularly the acknowledgment that the manuscript has not been examined recently or adequately. Rarely the very condensed form of description is not easily intelligible; cf. page 148, line 27: ‘Abbreviations. Few but extended.’ Also page 161, line 24 and elsewhere: ‘Abbreviations. Extended. Many.’ I assume ‘extended’ to mean that the regular list of abbreviations was somewhat enlarged in the manuscript being described.
Excellent indices of manuscripts, plates, authors and works, and a gen- eral index complete the volume. These serve also to make the companion volume of facsimiles more usable. For such a work these are matchless in number and excellence and enable the student to follow every step in the development of the Script of Tours.
In the narrow compass of a review only a few of the many matters treated can be even mentioned. In every chapter there are sections that elicit immediate approval, or cause one to question views previously held. Of considerable importance to scholars are the frequent indications of parts of the field of study that would prove fruitful subjects of investigation. Acquaintance with this book will be quite as helpful to the student as to the trained palaeographer.
The discussion of Alcuin’s influence on the Script of Tours (pp. 38 ff.) may be singled out as an example of Rand’s method of work and of the inti- mate knowledge of the subject given by years of study. Perhaps one could assume that Alcuin would have headed a reaction against the cursive hand, but even more characteristic of Tours is the careful individual development of the different script forms, capitals, uncials, half-uncials, and minuscules, and the complete avoidance of the mixed types, so common elsewhere in France at the time. On pages 23 and 24 there is a most interesting explana- tion of Q VII R as ‘q{uaternio] VII rfequisitum est].’ This custom of having an abler scholar recollate the manuscript is traced back, as far as concerns Tours, to Alcuin. On page 20, he notes that No. 140 alone has reclamantes; cf. page 166, where this manuscript is dated eighth or ninth century, fol- lowing Wordsworth and White. The presence of reclamantes in this manu- script alone at Tours is strange and deserves further consideration. Thomp- son (Introduction to Greek and Latin Palaeography, p. 54) says that ‘rec-
332 Reriews
lamantes to connect the quires first appear, but rarely, in the eleventh; from the twelfth century on they become common.’ Thompson’s acquaint- ance with Irish and English manuscripts is probably sufficient to make his statement authoritative for those styles of writing. Therefore it does not seem that the presence of reclamantes in No. 140 is explained by referring the manuscript to an Irish scribe. I may add that I have frequently found reclamantes in manuscripts of the works of Beatus from the tenth to the twelfth century. Of these Madrid, Bibl. Nac. H. 58 seems to be not later than early tenth century. There can be little question that the system was known in Spain from the ninth century on, and it may have been used in one of the original manuscripts of Beatus, 776 to 786. It would be interest- ing to have information on this question from early manuscripts of other parts of Europe, but an examination of early Visigothic manuscripts would, I think, prove most fruitful. The resemblance of the Ashburnham Penta- teuch miniatures to early Spanish art perhaps indicates an early connection between Spain and Tours. The reclamantes may be another hint of this connection.
Misprints or infelicitous forms of expression are most rare. I have noted only the following: page 72, line 5, ‘as examples as’; page 74, line 29, ‘St Maurice’s’; the Cajhedral of St Maurice is the earlier name (see pp. 3 to 58), but thereafter St Gatian is the name used; page 106, line 33, “No. 28’; as MS. 28 is being described, the comparison should be to another of the group, Nos. 24 to 34; page 204, line 17, probably an asterisk was omitted before ‘Fragment,’ for Rand was allowed to inspect this fragment.
Henry A. SANDERS, American Academy in Rome
EpGar Fin.tey SHANNON, Chaucer and the Roman Poets. Cambridge: Harvard University Press, 1929. Harvard Studies in Comparative Literature, vol. vu. Pp. xxii + 401.
THE gain in our knowledge of Chaucer’s indebtedness to the Classics of Rome may be indicated by a comparison of some forty pages in Professor Lounsbury’s Studies and the extent of the present work. The cautionary note sounded by Lounsbury still has value for the hunter of parallels: ‘Whether [Chaucer] was actually familiar with the writings of all of these [poets], or with all the writings of any one of these, is a question that cannot be answered positively, in the present state of our knowledge.’+ But Pro- fessor Shannon has shown, on the whole, a salutary fear of pressing too hard the arguments for some particular source. As we should expect, the book is thorough, sensible, and steady. Lounsbury’s further remark, that ‘it is not necessarily [Chaucer’s] tastes that are represented, but his op-
1 T. R. Lounsbury, Studies in Chaucer, etc. (New York, 1892), 11, 249.
ne A ee eels | ~ - n ; en en ee?! en. | qe sl ClCH et
— fp Ss = © SS of ee oe
> =| = ket
Reviews 333
portunities,’ ' leads us to speculate on the validity of making any inferences regarding the nature of the poet’s reading. But looking over carefully what Shannon presents compels us anew to wonder at the range and variety of what the busy fourteenth-century Englishman was able to assimilate. We are almost tempted to suppose that in beginning to write Chaucer sur- rounded himself with useful books like a modern investigator. Whatever his learning, he must have read with taste and insight; and a comparison in this respect with other men of the Middle Ages, like Chrétien de Troyes or Jean de Meun, would be interesting.
Mr Shannon does not attempt to offer a broad background for his study. He goes simply and directly through the poet’s works, one after another, and lists and summarizes the points of significance for his subject. In the second part he deals with the influence of certain Roman authors for whom there was no place earlier for adequate discussion. “The one dominant idea that emerges from the consideration’ is Chaucer’s ‘intimate knowledge of even the details of Ovid’s poetry.’? ‘The points of similarity in taste and poetic gifts between Chaucer and Ovid are so many that we cannot wonder at the extent of the Roman’s influence upon the Englishman.’* The con- geniality of spirit and temperament of the two writers is discussed at length, and the conclusion of the book is reached in the theory that it was the Italian Renaissance, chiefly Latin in its inspiration, which led to Chaucer’s freedom and mastery in his art,* — a conclusion of first-rate im- portance if it is justified, and in the light of the present book certainly one worth examination.
The difficulty, however, of weighing such arguments is that necessarily in such a field all the evidence is not yet at hand. Since the book first ap- peared, Mr S. B. Meech has made it clear that, where Shannon supposed that Chaucer was using a Latin version of the Heroides, his source was really an Italian translation; and Mr Meech promises to show that the poet also relied on the Ovide Moralisé more extensively than has been realized hitherto.® It is impossible to be quite sure that in every case where a Latin reference appears in the Middle English the borrowing has not come at second hand through a French intermediary. Thus doubtful cases play a large part in the estimate of Ovid’s influence, and we can rarely be certain whether Chaucer has taken his allusion directly from the Classics or from Boccaccio, Benoit, or the Roman de la Rose. The total impression gained from the treatment of certain poems, like the Troilus, may be seriously misleading, so far as Chaucer’s own acquaintance with the Latin poets is concerned; the analysis of Criseyde, for example, remains unconvincing as
1 Ibid. 2 Page 371. 3 Ibid. 4 Pages xiv, 376 ff. 5 PMLA., xxv (1930), 110 ff. and 117, n. 28.
334 Reviews
it now stands.' Furthermore, the question of emphasis, always essential, raises some doubts in one’s mind, as in the discussion of the House of Fame, where, in the opinion of the reviewer, Mr Shannon does far less than justice to French influence in his zeal to explain the material otherwise. He writes: ‘Chaucer, I believe, was carried so far away from his love-vision idea by his classical material that when he attempted to return to it, he found the spirit of the poem so changed that he gave it up.’ Not many readers, I am afraid, will agree with this view. The neglect of proper consideration of the Panthére d’ Amours and certain other works * leads one to a false idea of the whole poem and its general nature.
One returns dubiously, therefore, to Mr Shannon’s main point regarding the influence of the Renaissance. After all, if the House of Fame shows ‘imaginative freedom,’ we are not inclined to attribute that quality to the effect of the classics. If Boccaccio represents influence of the kind, he de- rived his stories from what must be considered a largely unadulterated mediaeval tradition. Furthermore, poets like Ovid and Virgil were well known long before the Italian Renaissance, unless we date that movement so early that it loses significance. Chaucer’s humane and realistic attitude toward his fellow creatures derived much from works like the Roman de la Rose and the Miroir de Mariage and considerable from his own tempera- ment. He employed a schematic presentation of his material until late in his career, and his free use of such a plan at its best (in comparison with the manner of a poet like Jean de Meun or the author of Piers Plowman) springs from his genius. Many writers in the Middle Ages could tell a story well. In fact Shannon seems to incline somewhat to an obsolete conception of the period. Thus he writes: Chaucer’s ‘spirit no doubt kindled in sympathy with the Latin poet’s spirit of fun, which so completely escaped the Middle Ages generally’; * ‘When the Renaissance came, and men began again to read the classics for themselves in a critical way instead of Scholia upon them, they noticed,’ etc.® But the value of the book is not affected by con- tentions of this kind. Rather it lies in the rich and varied detail ° offered to demonstrate the influence of the Roman poets in Chaucer’s reading, among
1 Shannon, pp. 160 ff. Cf. p. 167, n. 3.
2 Ibid., p. 117. Cf. the discussion of the Compleynt of Anelida, pp. 28ff., and Miss Fabin’s note, MLN., xxx1v (1919), 266 ff.
3 In addition to Sypherd’s material, cf. my note, MIN., xxx1v (1919), 323 ff.
* Page xvi. 5 Page 27.
6 Cf., for a doubtful instance of borrowing, p. 178, on Cant. Tales, A. 1761, and see Griffin and Myrick, The Filostrato of Giovanni Boccaccio (Philadelphia, 1929), pp. 64 f., n. 2. A few other details may receive brief comment: The discussion of the Troilus (Shannon, pp. 169 ff.) exaggerates, I think, the ‘dismay’ with which it may have been received. The question is one of proportion, like the poet’s ‘quarrel’ with Gower. Cf. the note on Chaucer as a student at the Inner Temple (p. 174) with Tout, Specutum, rv (1929), 382 ff.
Oo DP wes & s&s = Be na a fm. o& me.32 & & wee .-@
QO memo So
os
Reviews 335
which it is of special interest to find that Catullus seems to occupy a place. For generalizations, such as might tell us how far the English poet used the original Latin and how far he relied on the assistance of French or Italian translations, or what works Chaucer knew thoroughly as a student and what he picked up later, we must be patient enough to wait for other studies. Meanwhile here is sufficient material to provoke thought as well as gratitude, and to encourage further research.
Howarp R. Patcna, Smith College
MarcareET R. Toynsee, St Louis of Toulouse and the Process of Canonisation in the Four- teenth Century. Manchester: University Press, 1929. Cloth, 80. Pp. ix + 266.
THIs is no mere experiment in hagiography. The writer has deep and sym- pathetic interest in the career of the saint with whom she deals and makes clear all the available facts of his life and not simply those justifying the canonisation accorded finally to him by his friend, Pope John XXII. Transcending this initial purpose, she succeeds in giving a good understand- ing of the way in which, up to the fourteenth century, the Church managed the processes of canonisation in general. Competent theologians and stu- dents of ecclesiastical administration must pass judgment upon the details of her statements; to the lay mind