Imágenes de páginas

Bollandists. D'Alençon refers to this as R. Naturally the variants in R are of no real consequence and of no independent value, because, so far as they differ from A, they are the result of misreadings of A.

6. PERUGIA (M 69) 1046, the MS. used by Delorme for his « Legenda Antiqua S. Francisci ». I propose to refer to this as Per. 7. WORCESTER F. 75, to which I shall refer as W. Summarising for convenience of reference:

[blocks in formation]

Al. =

Assisi 686.

Marseilles (Boncompagni).
Poppi 13.

S. Isidore 1/73.


Perugia 1046.

Worcester F. 75.

The edition of P. Ed. Alenconiensis.

I know of no other MSS. containing the whole of or actual extracts from II Celano ('). That is not to say that fresh MSS. will not come to light, just as Per. and W have appeared. Leaving aside W and R (because R is merely based on a copy), we have texts AM PL and Per., all containing the whole of, or extracts from, the Second Life of II Celano. A detailed comparison of the texts and an examination of the variants lead to the conclusion that the problem of their inter-relation is a very complicated one. I think d'Alençon was wise in adopting A as his basis. It contains several substantial passages which are omitted by M. But it is far from perfect. It has omissions and errors, which can quite legitimately be. corrected by comparison with M. There can be little doubt that A and M, while differing in detail, are so alike as to indicate an ultimate common source. The problem grows in complexity when L and Per. are brought into consideration, for their variants show agreement sometimes with A, sometimes with M, but sometimes with neither. I will deal with this in some detail later.

We come now to our main subject W and its relation to the whole problem of II Celano.

W is contained in folios 128 r-145 r. It is not the whole of II Celano, but it is a substantial part. II Celano, as divided by d'Alençon, contains 224 Chapters or Sections. W has chapters or sections corresponding to 74 of these in other words it represents

(1) In saying this, I am leaving on one side the controversial question of the relation of the Speculum Perfectionis to II Celano.

less than one third of the whole narrative. This raises at once the question: what is the status of W? Is W. an abbreviated version of the original narrative known as II Celano? Or is W one of the sources placed under contribution by Thomas of Celano, from which he derived his material? While this second hypothesis is a very attractive one and while it would be very pleasant to believe that we have here to do with one of the primitive pre-Celano documents, I am forced by long examination of the document to the conclusion that this hypothesis is not consistent with the facts: that we have rather to do with a skilfully executed contraction or abbreviation of the original II Celano. In my view W is a text which is actually older in date than A or M or any of the other texts of II Celano, that it is nearer to the original than they are, and consequently of special value in settling the text.

Whatever the status of W may be, it is, however, not an original. It is contained in a composite MS., including many diverse documents. It contains gaps in the text, showing that the scribe could not read accurately, or did not fully understand, the MS. from which he was copying. Apart from such gaps, it contains actual mistakes, both in the text and in the capital letters for which spaces were left. In other words: W is a copy of the work of the abbreviator - not the original work of the abbreviator.

I come now to the examination of the W text in relation to the other extant texts.

[blocks in formation]

I note the following 15 instances in which W agrees with A and in which d'Alençon (= Al.) finds it desirable to emend the A reading:

P. 176, 1. 12 colligit.


1. 28 omission of oculis [a highly significant agreement, which proves relationship]. 190, 1. 25 scietis [in which W A and M agree].

[blocks in formation]
[ocr errors]

196, 1. 6 omission of et. 1. 10 Dei mei.

223, 1. 6 ambigens.

succedentes. 274, 1. 24 quoslibet.

[blocks in formation]
[blocks in formation]

251, 1. 28

278, 1. 22 omission of et. 287,

[merged small][merged small][ocr errors][merged small][merged small]

Then there are the following instances in which W fails to agree with A and does agree with d'Alençon's emended text which is, obviously, another way of saying agrees with M. These are really the most important of all, for W thus helps to settle the definitive text of II Celano.

[ocr errors]

[Note: Words or phrases omitted by A are asterisked.]

*a sensu.

1. 31 *eos.

[ocr errors]

177, 1. 3 ne. 180, 1. 22 tertio.

190, 1. 10 non verbis.

[blocks in formation]

181, 1. 22 *magni. 1. 25 dixi.

1. 24 ebdomada.

1. 12 *que. - 1. 13 quod. 1. 14 abscedet. 1. 17 conversi.



197, 1. 21

– 1. 23 timetis. — 1. 25 vos. — 1. 28 gratie. — 198, 1. 3 quos. — 199, 1. 1 proximo. 200, 1. 6 dies.

[merged small][merged small][merged small][merged small][merged small][merged small][ocr errors][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small]

262, 1. 26 intermixta. 263, 1. 11 spiritu. 264, 1. 15

fuit. 279, 1. 3 *periculis.

1. 29 *eo.

283, 1. 24 *inquit.

286, 1. 3 *nobili viro.

We have thus no less than 47 instances, in which confirmatory evidence for d'Alençon's emendation of the A Text is provided.

On the other hand there are a number of variants, in which W disagrees with A, and agrees with M, but in which d'Alençon preferred the A Text to the M Text.

[blocks in formation]

impleant. 226, 1. 6 petiit [M. petebat.] 240, 1. 1 completus. - 244,

[merged small][merged small][merged small][merged small][merged small][ocr errors][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small]

1. 5 peccata. Domino. 278, 1. 24 ideoque. 1. 29 non. 279, 1. 21 Domino. 309, 1. 27 cura. 321, 1. 1 fratres.


325, 1. 13 portari.

Relations of Lemmens' Extracts (L) with A, M and W.

An examination of the instances in which L varies from the A text shows that in the majority of cases the L variants are supported by the reading in W, thus indicating a distinct relationship between L and W. The following are the chief instances:

XXXI, 61. (217, 1. 2) Addition of comedere. · XXXII, 62. (217, 1. 29) Addition of laqueus. Ibid. (217, 1. 21) Substitution of Domini for Dei. - XXXV, 65. (220, 1. 14) Addition of simpliciter. [W simplex]. — CXV, 155. (286, 1. 4) Substitution of qui for quod. Ibid. (286, 1. 5) Addition of laesum. CXVIII, 159. (289, 1. 21) Omission of servus. CLVIII, 208. (325, 1. 14) Substitution of ipsa for ipso.

There are certain other L variants, which cannot be compared with W, because there happen to be no corresponding sections in W;

Archivum Franciscanum Historicum.



but if they are compared with M, they show a relationship of L to M. Thus:

LXXXVIII, 125. (265, 1. 12) Substitution of potissime for potissimum. - XCIV, 131. (270, 1. 8) Substitution of sic for sicque. — CIV, 143 (277, 1. 25) Addition of Domine.

More important, perhaps, than these minor verbal variants is the fact that in CIII, while A omits entirely section 142 (Non 80lum... meritorum) M provides that section, and L has an abbreviation of it. Unfortunately there is nothing in W to compare with CIII.

Again in CXVI, 157, the A text ends at benedicti, and M has a paragraph in quibus... reproborum', which L gives practically verbatim and which is also extant in Per. It may or may not be a coincidence that W also ends that chapter at benedicti.

There is one variant, the bearing of which is in the opposite direction, viz., against relationship of L with M. In LVIII, 91. (238, 1. 11) A, L, W, agree in the sentence: Credo equidem quod magis inde placebit donum quam lectio. M has no such sentence view a chance omission.

in my

On the whole, however, I agree with Delorme's view that it appears that the author of the L extracts did use a text of II Celano of the M type.


Relations of Per. with A, M and W.

Delorme observes in a footnote (1) that the anonymous compiler of Per. worked on a MS. of II Celano of the M type. I have examined this observation in some detail. I accept as a basis Delorme's view that Per. contains 24 chapters which are derived from or based upon the corresponding chapters of II Celano, as we know it; and that it contains a much larger number of chapters which are themselves the source or a source of the corresponding chapters of II Celano. I have only taken into account the chapters in Per. which are derived from II Celano. I notice the following variants:


XXXII, 62 (217, 1. 21) Substitution of Domini for Dei. [No evidence that M agrees, but L does]. XXXIX, 69. (223, 1. 24) Fratri meo inLXXIX, 113. (256, 1. 11) Verbis verba

stead of nostro. [M agrees].

instead of verba verbis. [No evidence that M agrees].

LXXIX, 113.

(256, 1. 16) After at ille addition of tu, inquit. [M agrees]. 113. (256, 1. 17) Eiecisti instead of egisti. [No evidence that M agrees, but W has iecisti]. LXXIX, 113. (256, 1. 22) Proferre instead of perferre. [No evidence that M agrees, but W does]. — CVII, 146. (279, 1. 10)

[blocks in formation]


Et monstravit instead of ac m. [No evidence that M agrees, but W does]. CXV, 155. (285, 1. 24) A Domino instead of A Deo. [M agrees]. CXVIII, 159. (289, 1. 25) Addition of et before operaris [No evidence that M agrees, but W does]. CXVIII, 159. (289, 1. 21) Omission of servus. [No evidence that M agrees, but W does]. CXLI, 188. (309, 1. 27) a sua cura instead of a sua custodia. [W and M both agree]. CXLVII, 195. (316, 1. 15) aliquando predicationibus instead of p. a. [No evidence that M agrees, but CLVIII, 208. (325, 1. 13) portari instead of portare. [M and

W does].

W agree].
W agree].

[ocr errors]

CLVIII, 208. (325, 1. 24) signis instead of signisque. [M and

These variants, some of very slight importance, serve to show the complexity of the problem of relationship. Out of 13 variants, we get 6 in which Per.'s alteration is supported by M, while in 3 there is the added support of W: there are 5 others in which M apparently does not support, but W does. I infer that if there is evidence, as Delorme thinks, of relationship between Per. and M, there is rather more evidence of relationship between Per. and W as Delorme would probably agree if he had seen W.

Summing up this question of relationship: I consider that M, L and Per. form a fairly closely allied group, while A on the whole the best text stands isolated, except for R which is dependent upon it. W is interesting, because it cannot be assigned absolutely to either group, because it shares idiosyncrasies with A on the one hand and with the M group on the other. We must hypothecate a prototype from which A, M, Per. and L all ultimately descend; and W appears to stand somewhere between that prototype and the original work of Thomas of Celano.

It will be observed that in the case of the two opening sections of the Worcester text, the full text of II Celano the version as published in Al. has been printed in parallel columns, so as to give the reader some idea of the extent of the abbreviation in W. The principal differences between the two texts are shown by italics. In Chapters 3 and 4 of W italics have again been used to indicate the variants from the Al. text. The general impression which the reader is likely to gather, is the skilful way in which the abbreviation has been executed.

It remains to say something about the Worcester MS. F. 75 and its probable date.

First of all it is to be noted that the Worcester Library is Benedictine in origin, a fact which helps to account for the survival of this Celano text.

« AnteriorContinuar »