but certainly not room for ten, impulsively, but providentially, made the blunder which the vigilance of the principal reviser has corrected for us. (5) But, indeed, he seems to have neglected, in like manner with the final UIT of the 'Indignos nos,' the last syllable of each of the two previous prayers, and that of the prayer which follows. In each, that is to say, of the four cases he, on nearing the end of a line of his own, dropped a last syllable which in the exemplar must have been detached from its context and lodged apart. Page ix. of the exemplar affords proof of my theory. The five lines in our volume, beginning with the sixteenth of fol. 9 v. and ending with the twentieth, resolve themselves easily enough into ten short uncial lines. They contain 184 letters. But when we come to fol. 10 we find that something has gone wrong. The content of the second and third lines is no multiple of eighteen or nineteen letters; and the fourth and fifth comprise ninety-one instead of seventy-three or thereabout. Why is this? The memory of our transcriber's behaviour in moments of arrested industry suggests the answer. No missal of St Gregory's can reasonably be supposed to have indicated an Antiphona for the Fourth Sunday of Advent. Our 'A. Memento nostri domine' must therefore be regarded as an adventitious usurper of the place of a suppressed UACAT. It was this change-a change made, I presume, in order to bring his work 'up to date'-that disturbed our mercurial artist, and caused him to make his lines, first too light, and then too heavy, till good luck set things right at last. This happened at the end of the Oratio. The latter half of Page ix., comprising, like the first, 184 letters, must have been somewhat as follows: TURUM PER DOMINICA EXCITADNEPOTENTIAM TUAMETUENIETMAGNA NOBISUIRTUTESUCCURRE UTAUXILIUMGRATIAETU AEQUODNOSTRAPECCATA PRAEPEDIUNTINDULGEN TIATUAEPROPITIATIO On the whole, then, and upon as careful a review as I have been able to give to my argument, I unhesitatingly conclude that the exemplar of the Corpus MS. was a volume of unicolumnar pages; that each page held twenty lines; and that each line had the average content of about nineteen letters. That the rulings were not all of absolutely the same width, is more than possible; for I find that Pages v. and vi. had 370 and 363 letters respectively, independent of capitals outside the ruling; that vii. and viii. had 373 and 390 respectively; and that ix. and x. had 368 and 382. Page xi. of the exemplar began with the fifth letter of 'natiuitatis' in the Secreta for Christmas-Eve (fol. 10 v., lin. 10). Counting thence to the end of the second Christmas mass, but omitting marginated capitals, and resolving all contractions with the exception of OMPS and of DNS, DS, IHC and XPC and their cases, I find that the total number of letters is 2283 (= 6 × 380})', a number sufficient to fill a hundred and twenty lines of the average content of 19 letters and an infinitesimally small fraction. This goes to prove that in the exemplar of the Corpus MS. the second Christmas mass ended at the foot of the verso of a leaf, and affords an obvious explanation of the notable display of artistic effort which marks the opening of the third Christmas mass in our volume. As to the question, then, with which I opened the present chapter, thus much, at least, is evident; that the exemplar of the Corpus MS. may have been the very book-or, rather, one of the two or more books -which St Augustine brought to England. It, manifestly, was a book intolerant of contractions, save the few which are known to have been in use in the age of Gregory the Great; and the very shortness of the lines is, I apprehend, sufficient proof that the script was uncial. THE CONSTITUENT TEXT OF THE TWO PROPRIA. The arrangement by which the first nine masses of the prototype of the Canterbury missals were made to fill precisely eight such leaves as went to the making of the exemplar of the Corpus MS. was not effected without several important changes in their constituent, and some little management, in their verbal text. I have already explained that a leaf now wanting to the Corpus MS. once held the re-written text of the mass for Advent Sunday, together with its Epistle and Gospel and the several portions of its officium. The 1 The numbers are:-for the second part of fol. 10 v. 405; for fol. 11, 750; for fol. 11 v. 726; for the former part of fol. 12, 402; their sum is 2283 = 120 × 191. The aggregate from ‘-tura solennitas' (fol. 8 v. line 16) to 'Da nobis dñe ut nati-' (fol. 10 v. line 10) is 370+363 +373 +390+368+382=2246=120 × 1818. loss of that leaf cannot be too grievously deplored, for it has involved the loss of information not otherwise to be had concerning the constituents of the mass; and the relentless scraping of our present seventh leaf, though it failed to obliterate the stain of the pigments employed for the opening words of the Oratio and for the initials of Secreta and Preface, did, unhappily, carry off both the rubric and the first letter of the Postcommunion. But, after calculating as best I can what must have been the textual value of the Preface, and applying the severest numerical tests at my command, I am convinced that the final constituent was longer by a half than that extant in the reprints; whilst, as to the Oratio, the existing condition of the vellum affords an all too cruel witness of our loss, for the prayer must have been almost as long again as its presumable precursor. The words 'Excita dne qs' are all that survives of it, but what followed it is impossible to guess. As to the Postcommunion, the case is not quite so deplorable. For, assuming, as we almost certainly may, that the erased Preface was that found in Ménard and Da Rocca, in Pamelius and in Muratori, 'Cui proprium est et singulare,' a composition containing 405 letters; and, knowing, as we do, that the Preface and Postcommunion together filled fifteen lines and a half of the transcript, or about (15 x 38) 589 letters, we may feel morally sure that the latter constituent was one of the four following, 'Concede q. o. d. hanc gratiam,' &c., 'Praeveniat nos q. o. d. tua gratia semper,' &c., Praecinge q. d. d.,' &c., or 'Fac nos q. d. d.,' &c.' The Secretae for the Second Sunday, for the Friday and Saturday in the Ember-week, and for Christmas-Eve are peculiar to the Corpus MS.; whilst Christmas-Eve and the day-break of Christmas have Prefaces, which, not having been cancelled by the owners of the book, must be regarded and treated as authentic. Another remarkable fact is the great length of some of the antiphonarial indications inserted after the titles of the several masses. Beginning with that for the Second Sunday, which must have had about 38 letters, I find that the aggregate number of letters in these indications was about 177; though the usual average would have yielded, I should say, 140. We may fairly say that they are 38 in excess of the normal number. My reason for mentioning this detail will appear presently. As in a later chapter I shall have to give some special attention to our ember masses for the summer season, I will say no more about them now. Excepting them, therefore, for the moment from more careful 1. See Migne, LXXVIII. 195, 196. consideration, I observe that, besides the eight changes just noted in the first nine masses of the book, our Proprium de Tempore has but three other instances of divergence from the constituent text of Pamelius and Muratori. They are the 'Pro Populo' for the Saturday after AshWednesday, and the Secreta and Postcommunion for the Eighteenth Sunday after the octave of Pentecost. But, when I examine these eleven instances I find that in no fewer than seven of them we agree with Ménard and Da Rocca. The accordance, moreover, is absolute; the Christmas Preface appearing in the curtailed form found by those editors, not in the longer and, presumably, original form proper to the Verona book (XL. viii.). A coincidence so striking would seem to discredit the theory that the manuscripts on which Ménard and Da Rocca worked exhibit a gratuitous and inexplicable succession of spurious variations capriciously foisted into genuine Gregorian work; and serves to confirm an opinion which I have long entertained, that we have (A) in Ménard and Da Rocca a first and perhaps tentative coordination of constituents and (B) in Pamelius and Muratori a new arrangement. The record of isolated instances yielded by the Proprium Sanctorum is very slight. We differ from Muratori and Pamelius once on the Vigil of SS. Peter and Paul, and once on the Feast of SS. Cornelius and Cyprian; whilst our mass for the Feast of St Caecilia differs from Muratori and Pamelius in the Postcommunion. There are, however, three complex groups or systems of constituent changes analogous to the two groups or systems just indicated in the Proprium de Tempore; and on these I shall have to dwell at the proper moment. When all shall have been examined, it will, I feel assured, be evident that the Corpus MS., besides its claim to exhibit a revision hitherto unsuspected of the verbal text of the Gregorian Sacramentary, and a structural text peculiar to itself, has established the further claim of embodying a new assortment of constituent elements indicative of a comparatively late recension. PROTOTYPE AND EXEMPLAR. Let us, then, by Redaction A understand the redaction to which is referable so much as is authentic in the documents made known by Ménard and Da Rocca, and by Redaction B that to which must be referred so much as is authentic in those made known to us by Pamelius and Muratori. The parent of the missals which underlie the Pio Clementine, and to which the Azevedian missal would seem to be referable, may be notified as Redaction C. By Redaction D I understand that outcome of editorial effort from which, as from its proper source, was derived the liber missalis which Augustine and his monks brought with them to our shores in the year 597. For I believe St Augustine's liber missalis to have been a modification of that document. It is obvious that the differences by which a later edition of a work is distinguished from an earlier may be introduced into the document in the course of a review prior to transcription, or in the course of the transcription itself. But it is obvious that, even though the editorial achievement be perfect, sufficient, satisfactory of the editor's full intention, the editor himself may, after the new archetype has left his hands, see fit to call it back in order to introduce into it some change or changes too specific in themselves and too limited in their scope to justify him in calling the resultant by the name of a new edition. Hence my reason for speaking of the Canterbury original as a sub-redaction of the prototype which for convenience' sake I denominate by the letter D. For, curiously enough, the several groups of neighbouring prayers of which I spoke in my last chapter1 as constituting an important difference between the constituent text of the Corpus MS. and that of sacramentaries of the Pamelian type have, one and all of them, a stichometrical characteristic which goes to prove that they are the outcome of a manipulation of the prototype after the prototype had issued fresh in its charms of careful script and comely rubrication from the papal scriptorium. Let us begin, then, with the group of changes comprised in the first nine masses of the book, and tabulate their textual value in terms of letters. The substituted Oratio and Postcommunion in the mass for Advent Sunday yield an increment of some 170 letters; on ChristmasEve a Secreta of 104 letters is replaced by one of 183; whilst two new Prefaces contribute between them new text of the value of 435 letters. The aggregate of these augmentations approximates so closely to the double of a figure already made familiar to us that our curiosity and interest are aroused, and we bethink ourselves of the twice nineteen letters in excess of the average yielded by the antiphonal indications* and of the 24 letters of the clause 'aduenienti,' &c.; and we find, to our mingled amazement and delight, that the result is as follows: 1 See above, pp. cxv., cxvi. 2 See above, p. cxvi. 3 See above, pp. lvi., lxxxiv. |