SOURCE: Allen, Hope Emily. “The Origin of the Ancren Riwle.” PMLA 33, no. 3 (September 1918): 474-546.

[In the following excerpt, Allen presents and discusses evidence that Ancrene Wisse was written not long after 1134 in the hermitage of Kilburn.]

The following paper will give a preliminary statement of a new conjecture as to the origin of the Ancren Riwle.¹ It is proposed to identify the three maidens for whom the treatise was composed with the “tribus puellis, Emmae, videlicet, et Gunildae et Cristinae,” to whom, according to the charter printed by Dugdale,² the hermitage of Kilburn, with its appurtenances, was granted by the Abbot and convent of Westminster sometime between the years 1127 and 1135.³ The hermitage was endowed permanently with money, land and beneficia, in return for which the inmates were to be the beadsmen of the abbey and of its confederate, the Abbey of Fécamp. The house at Kilburn was to be under the protection of St. Peter's, but it was to have complete independence in regard to its internal affairs. The establishment thus made had a continuous existence till the Reformation, under the title, which it seems to have acquired very early, of “Kilburn Priory.”⁴

The records of Kilburn Priory show that striking likenesses existed between that house and the establishment of anchoresses for whom the Ancren Riwle was written. These, with some evidence confirming the connection of the treatise with Kilburn, will be enumerated point by point.

(I)

The Ancren Riwle was written for women who, like the recipients of the hermitage of Kilburn, were three in number. They were of noble birth, and young, and their entrance into religion had occasioned a stir in the world.⁵ The youth of the Kilburn ladies seems implied in the use of the word “puellae,” which is applied to them in the charter of foundation, and their noble birth is put beyond all doubt by a statement of John Flete, the fifteenth century prior of Westminster (who, as such, would be, as we shall see, infra, p. 495, in constant connection with the house in his time, and thus in a position to know its traditions); he says that the “virgines” whom Abbot Herebert established at Kilburn were “domicellae camerae” of “good Queen Maud,” the wife of Henry I, daughter of St. Margaret of Scotland, and niece of King Edward the Confessor.⁶

(II)

The three women are in both cases inclusae. The references to this fact in the Ancren Riwle are continuous, from the first pages on. The fact that the Kilburn sisters are also enclosed appears from the use in the foundation charter of the phrase, “quae infra clausae fuerint,” as applied to the sisters' residence in the hermitage of Kilburn, and a plea is quoted from 1207 by Park (p. 188), in which reference is made to the Prioress of Kilburn, “quae inclusa est.”⁷ The Riwle makes it clear that the nuns addressed are enclosed in a church (Morton, pp. 68, 142, 242, 262), and the second Kilburn charter printed by Dugdale implies the same for the Kilburn sisters,—“quae sunt in ecclesia beati Johannis Baptistae de Keneburna.”⁸

The failure to recognize that the Kilburn sisters were “inclusae” is probably responsible for the failure of scholars heretofore to identify them with the ladies of the Riwle, when all the surface indications so plainly pointed the way to that identification. It has not been well understood that “inclusae” might be women living in a definite organization, with a superior. In this connection it is interesting to observe that once, in giving a more or less formal statement of the sisters' condition, the French text uses the term “recluse de moustier” (f. 67v), where the English gives “chirche ancre” (Morton, p. 416).⁹

(III)

We are told in the Ancren Riwle (p. 356) that the three sisters are “beadsmen”; the Kilburn sisters are the beadsmen of Westminster Abbey, as had been the hermit Godwyn, the builder and original occupant of the hermitage of Kilburn. It may be observed that traces of the duties of prayer expected of the anchoresses of the Riwle in return for their sustenance appear in the regulations for their service. Prayers for the dead make an important part of their daily routine: it would be well if they were to say Placebo, Dirige, and Commendacion every day (Morton, p. 22), though on the eve of a festival of nine lessons¹⁰ they may say only three lessons of the service for the dead. … “The anniversaries of your dear friends” are probably the anniversaries of Westminster, of which a list from a later period is printed in the Customary of the Abbey (p. xxvii).¹¹

(IV)

When the author of the Riwle describes the material circumstances of the three anchoresses, he says that they enjoy a very unusual security in this respect: many others suffer great want.¹² These statements make it all the more significant that the nuns of Kilburn enjoy just such material security as is described in the Riwle. Moreover the assistance which the nuns of Kilburn receive, like that described in the Riwle, is not only secure, but perfectly definite and regular. It is one of the most striking correspondences between the two lines of evidence which we are following that the corrodies received at Kilburn seem specifically referred to in the Riwle. … It would appear that the “hall of the friend” was the hall of the Abbot of Westminster, who had granted the ladies of Kilburn by their foundation charter a sum of money and “duo beneficia” (“one for the benefit of the souls of Westminster and Fécamp,” and one “which Ailmar the hermit had”); very soon a second rental is granted them by the Abbot (one was given by a priest at the foundation), and their other benefits are soon described as “three corrodies,” of which one is that “which Ailmar the hermit had.”¹³ A vagueness of expression makes it impossible to decide just the relation between the benefits of Kilburn described in the charters of the first two abbots concerned with the house, but it seems very likely that the ladies may have been receiving the three corrodies before 1138, which was approximately the time of Gervase's accession. No sign is given in the charters as to any increase in the personnel of the hermitage, and if there were none, each of the three ladies would be receiving by 1138, if not before, from the hall of the patron of her house her “bread and what is eaten with bread,” just as did the sisters of the anchorage. Her clothes and the food of the serving maidens (which were also secured for the women of the anchorage) could be provided out of the rentals of Kilburn, even if the eleemosynary sum given at the foundation had been discontinued.¹⁴ Since the source of all their support was the abbot, he might very easily be the “dearest friend” to whom the women of the Ancren Riwle were to show their need if such arose (p. 416).¹⁵

The Ancren Riwle gives us hints as to the fare received by the sisters, which would seem to suggest the monastic food supplied by corrodies. The anchoress is not to complain of “mistrum, oðer leane mel, of unsauure metes, of poure pitaunce” (Morton, p. 262). The French text at this point speaks of “defalte de repast ou de sauourees viandes ou de pure (?) pitance” (f. 40v).¹⁶

In contrast to the fare of Kilburn may be cited the example of Tarente, the nunnery with which the Ancren Riwle was formerly connected (see infra, p. 542). This received from the hall of the lord of the manor, who was its patron, a tithe of all bread “except the king's bread, and all the tithe of salt pork, and of cattle killed in his house every year.”¹⁷

… All the way through the text the author has so combined admonitions directly intended for the three sisters with what he expressly tells them is not needed for them, but may be useful to others,¹⁸ that in a case like the present it is difficult to separate general references from those that reflect actual conditions in the life of the women in question. At the same time, considering all the other circumstances, it would appear that this passage may be taken as an admonition adapted to the special circumstances of the women addressed,—given impersonally, with a slight touch of affectionate sarcasm. There would be nothing startling in the author's referring to the sisters as “rich,” since he has already (p. 192) spoken of their material circumstances as superior to those of any anchoresses whom he knows. B, in any case, here uses “church” instead of “rich” (Macaulay, p. 330). As a matter of fact, the sisters of Kilburn were “tillers of land and receivers of rents” from the first. They owned land in Southwark and Knightsbridge, and there was land connected with their hermitage.¹⁹

(V)

In the foundation charter of Kilburn it is stated that Godwyn is to continue to live there as “magister loci, illarumque puellarum quamdiu vixerit custos. Et post ejus obitum eligat conventus puellarum seniorem idoneum, qui earum ecclesiae praesit, abbatis tamen concilio.” The anchoresses of the Riwle are also under a “master.” …

B (apparently written for the house descended from the original anchorage) contains many references to “your master” (see Macaulay, pp. 470-2). It is of the greatest interest that one implies that the “master” is the author of the Riwle. …

The author's “mastership” would give the perfect explanation for the special authority, knowledge, and affection which he shows in relation to the sisters (who are “les femmes qe ieo plus eym,” f. 22v, p. 116). Godwyn probably knew his charges in their secular life, for it was at his “concessu” and “precatu” that they received their hermitage. No reader of the Ancren Riwle can doubt that its author would be, in the literal sense, “a fair for the mastrye.”²⁰

(VI)

The Kilburn church was dedicated to St. John the Baptist, and in a transaction of the reign of Henry IV the name of the Virgin is joined in the dedication (Park, app. p. xxx). About 1200 Kilburn is referred to as dedicated to St. Mary (v. infra, p. 490 n.). In the treatise the Virgin and St. John are taken as the types of the solitary life as described in the New Testament, and a specially lengthy discussion is given to St. John (Morton, p. 158). The whole account of the solitary life at this point is taken, with some minor changes, from the Carthusian Customs drawn up about 1127 by Prior Guigo;²¹ but this need not destroy the significance of the passage for the present discussion, for its original use by the Carthusians was exactly that which has been suggested in connection with Kilburn. Guigo's praise of the solitary life in the Customs, given with special reference to the Virgin and to St. John the Baptist, made the climax to the Carthusian Customs, because Carthusians were dedicated to the life of solitude almost as specially as anchorites, and their patrons,—to whom they made their vows on profession—were the Virgin and St. John the Baptist.²²

In all the versions and manuscripts of the Riwle the ninth day set for communion is June 24 (p. 412), the day of the patron of Kilburn, but all the manuscripts designate it as “Midsummer Day,” except the French, which calls it the “Day of St. John the Baptist” (f. 67)—the title which we should expect at Kilburn. The French form of this passage will be discussed later (p. 541), and will be seen to show other evidence of superior authority.

(VII)

When about 1140 two hermit women were taken under the protection of the Abbot of St. Alban's, and given companions and monastic buildings at Sopwell, we are expressly told, not only that additions were made to their numbers, but that the Abbot—although they are called “inclusæ”—“more sanctimonialium velatas, et sub Ordine S. Benedicti victuras constituit.” When about 1134 his neighbor the Abbot of Westminster had put three women into the hermitage of Kilburn, it is a contrast significant for the present hypothesis, that we are given no sign of the rule which they are to follow. This fact has excited the astonishment of Park (p. 162), and he is still more astonished that no mention of the rule followed at Kilburn is to be found till after the middle of the fourteenth century, and that in a formal document of 1377 the nuns are called “of the order of St. Augustine” (Park, pp. 171, 177). This information he is inclined to discard, because of the connection with the Benedictine house of Westminster, and, accordingly, for the same purely conjectural reason, the historians all call Kilburn Priory Benedictine. There exists, however, practically no mediæval evidence for this title, and the Augustinian connection is repeated in a Patent Roll of Edward III (p. 340),²³ and supported by the excellent authority of Prior John Flete, who, in the passage already quoted, calls Kilburn Priory a “cell of canonesses.”

The gradual process—hard to trace in the records—by which the Augustinian rule was attached to canons at just this period has been described by Dr. Frere,²⁴ and perhaps the same development went on in time in the case of irregular communities of women. In 1244 the Pope united some communities of hermits as the “Hermits of St. Augustine” (who thereupon made the claim that their manner of life had been kept alive since the days of St. Augustine). We do not know when the Augustinian Rule was first applied to Kilburn, and this may have been a precedent influential in the case of that house.²⁵ Since, at least in early days, Benedictine houses like Marcigny and Cluny counted anchorites and hermits among their numbers, it would appear strange, however, that Kilburn did not come into the order of its protector.²⁶

It is significant that the Rev. Vincent MacNabb, in an article to be discussed in the Appendix, has made the interesting discovery that the Rule of St. Augustine was used in the Ancren Riwle. It may be added that the sisters of the anchorage celebrate festivals of nine lessons, like Augustinian canons, secular clergy, and lay-folk, instead of festivals of twelve, as was the custom in the Benedictine order.²⁷ An Augustinian influence would be natural in the Riwle, if written for Kilburn, since, as Dr. Frere points out, Augustinian houses were the height of the fashion until the Cistercians began building in England, and conspicuous Augustinian establishments were made by Henry I and by “good Queen Maud.” The Augustinian Rule could be imposed on local customs, and old houses could thus get the prestige of being Augustinian without alteration in their habit of life. The Gilbertine order is a mirror of contemporary influences in that its nuns were Benedictine, its canons Augustinian, and its lay-brothers Cistercian.

(VIII)

The most surprising manuscript of the Ancren Riwle is B which, as we have already seen, is at once the earliest copy, the most correct in the minutiæ of the text, and the most interpolated. By their mention of friars, some of the new sections cannot have been composed much earlier than 1230, and by its handwriting B cannot have been much later.

It is clear from the additions themselves that they are directed towards a special community, and we seem to have good reason to believe that this is the house for which the rule was originally written. There are several cross-references to the original text as “your rule,” and a great effort is apparently made to coalesce the new material with the old. The community addressed is spoken of as being “like a mother house”—not, we should be careful to note, as being a mother house—and the author goes on, apparently as a metaphor for their extraordinary unity, to say that they are united, “as if they were a convent of London, Oxford, Shrewsbury or Chester” (Macaulay, p. 470).²⁸ He goes on to say that the fame of the community he addresses is everywhere known—especially for their extraordinary unity—and their “convent” is spread over England. This again seems to me to be a metaphor, and perhaps a reference to the wide dissemination of the Riwle, originally written for this house. That the additions were considered somewhat authoritative, appears from the fact that they were to some extent copied in some of the later manuscripts—and in the French and Latin versions—and a manuscript (C), already written, was at some points corrected to bring it into line with the version represented by B. It is significant that the place-names above quoted begin practically at Kilburn, and follow in the order of their relative distance from there.

Another fact of the greatest importance is added in the same passage, and furnishes the basis for the interpretation given above, and a very important link with Kilburn. We are told that the community addressed now number “twenty recluses or more”—the “most living together anywhere in England.” It can easily be seen how, from their tradition of isolation, twenty recluses living together in the same unity which the author of the Ancren Riwle had recommended for three—in the passage (p. 254) to which was later appended the section described above—would be worthy of special praise, and this would be the only case in which unity, the first of the conventual virtues, could be so considered.

The mention of the twenty recluses has another very important implication for the present hypothesis; it gives a satisfactory explanation for the omission from B and the later copies, of the description of the personal circumstances of the three sisters.

It is perhaps the strongest single piece of evidence for the identification that is here proposed—apart from the similarity in the personal circumstances of the original inclusœ—that the new material in B, added about the year 1230, can be connected with the history of Kilburn during the years 1225-31. During that period a great dissension arose between the Bishop of London and the Abbot of Westminster in regard to the jurisdiction over the cell of Kilburn—in spite of the fact that this had, at the foundation of the house, been formally secured to the abbot by Gilbert the Universal. This dispute probably made part of the general controversy as to the exemption from episcopal control which Westminster was undergoing at this time, and the particular question of Kilburn was settled in 1231 by a commission appointed by the Pope (who had been twice appealed to in connection with the cell). It is noteworthy that the convent yielded to a compromise on the subject of Kilburn, in spite of what had apparently been the custom from the beginning, though they fought out the larger issue to a successful conclusion.²⁹

The terms of the settlement made in 1231 as to the supervision of Kilburn would seem to show that some abuses had crept into the house from the too exclusive power exercised over it by the abbey. It is stipulated that no member of the convent except the abbot or, in his absence, the prior, shall have access to Kilburn to hear confession and enjoin penance, and the Bishop is to have the right of visitation, and of hearing even secret confession. As Park again notes with astonishment (p. 171), there is still no mention of a rule. The real power seems to be vested still in the successor of Godwyn—who appeared in the plea of 1207 as “magister et custos”—and, as it is important to note, is here described as the “secular priest who is set over the house.” A chapter house and a prioress are mentioned and would seem to imply a great increase of numbers,³⁰ but nothing definite on this point appears in the records until the latter part of the fourteenth century,³¹ and it is doubtful whether the inmates would ever have numbered more than twenty: thirteen was, as a matter of fact, the number to which their neighbors of Sopwell were restricted at the time of foundation,³² and it was the number in Carthusian houses.

Though the information given by B makes clear that in point of numbers the house in question may agree with what we know of Kilburn at the date of the writing of B, it cannot be said that B is quite explicit as to the manner of life there. It is evident that the anchoresses of B meet the outside world only at the windows of their cells, and that also their personal intercourse, as it were, with their sisters, is carried on by visits of their sisters' maidens (Macaulay, pp. 466, 473); but no mention is made of leaving the cell for religious service or for chapter, or of not leaving it. In any case, we know that in 1207 the Prioress of Kilburn is called “inclusa,” and the absence of any mention of a rule in the Kilburn records of 1231 may point to an unusual manner of life. They certainly give no information which would make it impossible that we have to do with anchoresses, unless it be that we are told that at the visitation of the bishop the prioress and nuns are to come in procession to the chapter-house, but we shall see later (p. 542) that B practically implies that the house addressed is a priory. It would be hard to see how a settlement of twenty recluses could keep together without some corporate life, for which a chapter would occasionally be necessary, and it would appear inevitable that the strictest degree of enclosure must in any case have to be abandoned in so large a group, since all could not have windows looking into the church, and they would therefore have to go there to hear mass. The original three sisters had windows from which they saw the altar (Morton, p. 68), but that does not make it impossible that they also attended mass in their church, and chapter, according to the example of the Carthusians.³³

We have seen that in the settlement of 1231 the most important innovation is the visitation of Kilburn by the Bishop. It is a striking coincidence that, as Mr. Macaulay has thought it “interesting to note,” “the precautions urged upon the anchoress with regard to shewing herself unveiled … are expressly extended to the case of a bishop's visit” (p. 463).³⁴ The passage in question (p. 466) apparently describes a bishop's visit to the anchoress's window, but such an incident would be quite possible at Kilburn according to the terms of the settlement of 1231. The bishop was then given the right of hearing “confessio privata,” which might have taken place in the cells, as was the Carthusian custom.³⁵ The fact that the writer of the new passages would doubtless be on the side of the abbot in the controversy perhaps accounts for the somewhat suspicious manner which he shows towards the bishop.

The causes which may have existed for curtailing the exclusive power of the abbey over the cell do not appear in the records at present available,³⁶ but it would appear from B that specific abuses have suggested the composition of the recension, and that they are such as would be very foreign to the original anchorage, but might arise from the connection of the house with the abbey. Whereas in the original Riwle the author continually reminds the sisters that they already live in some ways only too strictly, and do not need his counsel in respect to the grosser sins (see Morton, pp. 8, 50, 68, etc.), the author of the new directions seems to hint mournfully at irregularities, present or just past, among those whom he is addressing. Their conversations with men unveiled are restricted with the greatest anxiety, and they are told that this is the most important external regulation of their rule (p. 466). The text of Jeremiah on the darkening of the finest gold, which is the stock text used in the Middle Ages against degenerate religious, is commented on with great feeling, in a passage (p. 467) which accuses some anchoresses, at least, of extreme worldliness, bitterness, desire for prestige, and extreme love of dress. As a matter of fact there is no passage added by B which does not dilate eloquently on some sort of conduct most unsuitable for an anchoress. We can, therefore, hardly escape the conclusion that the recension of the Ancren Riwle which is present in B was written at a time of the reformation of the house, which would appear to be that of which the records of Kilburn tell, at just the same period. We do not, of course, know who was the master of Kilburn at this period, but we know that the abbot of Westminster, Richard Berking (the special counsellor of Henry III, who is supposed by Widmore [p. 42] to have influenced the king to rebuild the abbey) was a man of unusual power.

It would perhaps be suspected that we have in B an official copy—perhaps an original—preserved at Kilburn, but the absence of the name “St. John the Baptist,” which would naturally be used at Kilburn for the ninth date for communion, has already been mentioned, and this passage in the French text (which gives the Kilburn patron), and the French version of another, to be discussed in the Appendix, seem distinctly superior as a whole, to that given by any English manuscript. The same state of affairs appears in details; Mr. Macaulay has noted that the French text in general supports “what seem to be the original readings, as opposed to those of the manuscript followed by Morton” (p. 70). A trifling example that shows superiority over B is in the case of the passage from Jeremiah, already mentioned. B here has the absurd “sein ierome,” whereas the French has “li prophete Ieremie” (Macaulay, p. 467 n.).

If B cannot be considered to be the original of the second form of the work, it is also obvious that the French cannot. What seems to be true is that every manuscript, for one cause or another, stands very far from the original; B, because though evidently nearly related to an official copy, this copy was a second version; N because of its general carelessness, in spite of its superiority as the most complete representative of the original first version; the French, because though, unlike the others, it possibly is derived from the copy in the original language and in details often shows distinct superiority, yet it, like B, is related to the second version and is only an incomplete copy of that—as is perhaps natural, since it is to be dated a hundred years later than B. This state of affairs would seem to suggest a long circulation for the work, and it is useful to observe, in this connection, that Mr. Macaulay notes that “the reviser whose text is represented by N was in the matter of grammatical inflexions in some respects ‘earlier’ than B, especially in regard to pronouns, articles and demonstratives” (p. 149 n.), though, in general, the “text of N represents a fuller development,” which, however, is not “necessarily later in time” (p. 149). He notes (p. 65) that early and late forms are to be found mixed in the French text, and he finds no difficulty in assigning the original composition to a date a hundred years before that of the manuscript. The rapid changes in language found in prose works would probably make an even earlier date possible. It was noted by Heuser³⁷ that the Ancren Riwle shows a mixture of late and early forms. Again we find evidence that would seem to prove that the work had passed through a considerable period of circulation. Altogether, it seems very hard to believe, when the manuscripts and versions in 1230-50 show such variations of all sorts, that the original treatise could have been composed at the beginning of that century. It is one of the many important results of Mr. Macaulay's investigation of B that, by showing the serious alterations in the text which were existing at the time of the writing of our earliest copies, the date of the original composition of the Ancren Riwle is put back far beyond where it has usually been placed. More than that, B suggests that the house addressed is near London, and a priory—to go no further in summarizing its benefits.

(IX)

One possible indication of a date for the Riwle earlier than 1160 occurs in the statement by the author: “Eresie, God beo iðoncked, ne rixleð nout in Engelond” (Morton, p. 82). With this should be compared the description of the Waldenses who came to England in 1160 by William of Newburgh; he declares, “Sane ab hac et ab aliis pestibus hæreticis immunis semper exstitit Anglia, cum in aliis mundi partibus tot pullulaverint hæreses.” From the time of the Britons the “virus of heresy” “nec in eam aliunde usque ad tempora regis Henrici secundi tanquam propagandum et dilatandum introivit.” Since the Continent was at this time full of heretics—so that Peter the Venerable, so many times mentioned here, devoted to them much of his energy—the immunity of England would appear conspicuous. They went on increasing on the continent till the Albigensian Crusade of 1215, and it is hard to believe that they did not again reach England. A letter of Peter of Blois of c. 1191 seems to refer to heretics in the diocese of York, and Higden tells us that some Albigensians were burned in England in 1209.³⁸

(X)

Dr. Robinson believes that the foundation of Kilburn made one of three projects in which Osbert of Clare took a principal part. Osbert's relation to Kilburn seems to me doubtful, but it is at least certain that he took a leading part in the movements to canonize King Edward, and to establish the Feast of the Immaculate Conception of the Virgin. The latter enterprise met with opposition, though it was approved by a council of English bishops in 1129.³⁹ It is not certain that the celebration with which Osbert was connected took place at Westminster, for it falls at a time (1127) when he is supposed to have been in exile.⁴⁰

Father Bridgett has pointed out⁴¹ that we have an echo of this controversy in the statement of the author of the Riwle (p. 38) that there was no sin in the Mother of Christ after His Conception, “as I believe, whatever may have been before.” The author was here suspending judgment on the exact theological doctrine that underlay the festival.

(XI)

Even if the treatise were not originally written in English, it is quite clear from the English texts that it was written for persons knowing English, and therefore, in the conditions of those times, almost certainly of English blood. …

The French should also be quoted here:

“Quantqe vous vnques dites daltres celes prieres, sicome pater nostres, et auees, psalmes et oroisons en votre maniere demeine, ieo en sui bien paiez. Chescune die ausi come mieuz li aporte al queor, versiller en psalter, lire euangeiles (sic) ou en francoys seinte meditacions.”

(f. 8)

Though the divergence of the French text here may make a slight uncertainty as to this passage, it would appear that the English may give the original reading. The fact that the treatise was at least translated into English very early, would support the presumption that it was written in an environment where English was read. Moreover, there is a reference (Morton, p. 244) to the “English book of St. Margaret”—a very interesting indication of the use of English literature at this time—which is all the more interesting if the work in question can be identified with the alliterative life of that saint⁴² which makes part of the “Katherine-group” of legends judged to stand so near to the Ancren Riwle both in dialect and in spirit.⁴³ Altogether, it would appear that we were justified in taking it for granted that the Ancren Riwle was written in a distinctively English environment.

Archdeacon Pearce has remarked in his work on the monks of Westminster that “the fact appears to be that from its quite early years the Convent was predominantly English” (p. 36). Of the persons connected with the foundation of Kilburn, several were apparently natives. Godwyn, the Master, and “Ailman sacerdos,” the donor of the land in Southwark (noted in the first charter), would appear to be so by their names, and Osbert of Clare was an Englishman.⁴⁴ Flete tells us that Abbot Herebert was a Norman (perhaps, as Dr. Robinson conjectures, one who came from Bec with Abbot Crispin), but Widmore remarks that this must be “only by family, or at farthest by birth for he was a monk here, and almoner of the convent, when appointed abbot” (p. 23). He was the first monk of the house chosen abbot after the Conquest.

Not only is it certain that some of the special patrons of Kilburn at its foundation were Englishmen, but we would seem to have reason to believe that the three occupants of the hermitage were at least not all Norman, since one bore the name “Gunhilda.” As a matter of fact there would be some reason why all three names should be given to English women, for they all occur in the family of King Edward the Confessor. Emma was his mother, Gunhilda his half-sister (the daughter of Canute), and Christina his niece, the nun of Romsey who brought up Queen Matilda.⁴⁵ We know that some of the strength of Henry I's position came from his reconciliation of the English elements of the nation, and that this was specially furthered by his marriage with a princess of Anglo-Saxon royal blood. William of Malmesbury, in a passage often quoted, tells how the Norman courtiers called the king and queen by English nicknames, and this would seem to suggest English influences at court, and doubtless English persons in attendance there.

(XII)

Since Mr. Macaulay accepts the French text as the original version of the Ancren Riwle, the question of the date and locality of the English text may not be vitally important for the present hypothesis. However, even if the work were not first written in English, it is highly probable that it was soon translated, and at the original anchorage. The fact that the English text found in N retains the personal details of the work—which so many later copies omit—would seem to show that it was made before changes in the conditions described had made these details uninteresting. It would be likely that the place of the translation would be Kilburn itself, where the serving maidens—to whom the last section was to be read weekly—would, it might seem, need such a text,⁴⁶ and the example of the use of the vernacular for lay religious (sometime in the twelfth century) by the Cistercians—who were the most potent influence of the day—must be remembered in connection with the English text of the Riwle, as well as the long tradition for vernacular religious instruction in Anglo-Saxon England.

Since, therefore, it is highly probable that the English text, even if not the original, should be written at Kilburn,⁴⁷ it is important to localize the dialect of the English version. Mr. Macaulay, who is the only writer who has had the advantage of knowing all the manuscripts of the work, writes of them as follows: “B, C, and G constitute a group resembling one another closely in forms of language, and belonging to that particular development of the Southern dialect, on the borders of the Midland region, which is exemplified in the early lives of St. Katherine and St. Juliana,⁴⁸ the purest form of this appearing in B, which is also distinctly the earliest in time of our manuscripts. … T has as its basis a text of the same kind, but was evidently written by a North Midland scribe” (p. 148). The latter is the copy which, next to N, best preserves the original details of the anchorage. N, which, as we have seen, presents the original form of the text better than any other whatsoever, is “distinguished from all those that have been mentioned by features characteristic in this period of the purely Southwestern dialect” (p. 149). It is significant that the two copies most alien in dialect are those most out of touch with the later development of the text.

The facts just reviewed would make it appear that the composition of the English version took place in the Southern dialect region, on the border of the Midland, which was the home of three of the thirteenth-century manuscripts, and of the parent of the fourth. In exactly this region London was included at the time of the Proclamation of Henry III of 1258 (which is approximately that of the writing of our manuscripts), as an analysis of that document has shown;⁴⁹ though earlier it had doubtless spoken a pure Southern dialect, and later the influence of the neighboring Midlands and of its cosmopolitan position (already active in 1258) was to make its speech almost consistently Midland.⁵⁰

The age of the dialect of the Riwle would not appear to be a valid obstacle to the present hypothesis, even if the origin of the English version be put a very few years after that of the original (which would thus become very nearly the earliest piece of French prose extant). The date of the foundation of Kilburn is generally put, as we have seen, at 1134; but it is evident that the treatise is not written at once on the retirement of the three sisters, since their virtue as anchoresses seems to have been thoroughly tested at the time of writing, and the author remarks that they have often asked him for a rule. Therefore the composition of the work could probably be put several years after 1134. A long circulation before the writing of the extant copies would inevitably bring serious changes in the language, yet it would appear that we have several touchstones which may be used to test its date, though none perhaps is quite perfect. The Peterborough Chronicle, which is securely dated, belongs to another dialect, and is, moreover, deliberately archaising;⁵¹ most of the manuscripts written in the middle of the twelfth century contain transcripts of older works, and are therefore obviously no fairer standard for the speech of their own time,⁵² than the Ancren Riwle is for the speech of the early thirteenth century, if, in spite of the thirteenth-century manuscripts, it turns out to have been composed in the middle of the twelfth. Altogether, however, in spite of the lack of English literature accepted as composed in the first half of the twelfth century, the characteristics of late dialect found in the Riwle—so far as they can be observed without an exhaustive study and collation of the manuscripts—can be all found at times in the various manuscripts and legal documents which have been considered our only basis for the study of English of this period.⁵³

Though the dialect of the English Riwle can doubtless be fitted sufficiently well both to the time and place that is here proposed for it, yet this subject should not be concluded without an influence being mentioned which would give sufficient explanation for any deviations in the Riwle from the forms occasionally found in documents written in London at the time of the settlement of Kilburn. Local usage may sometimes war with convention in any documents of the time, with inconsistent results; for the use of a “Schriftsprache” seems proved for this period by the existence of West-Saxon legal documents written in the eleventh and twelfth centuries in the North.⁵⁴ With this fact should be connected the statement of Giraldus Cambrensis, of a slightly later date, where, after declaring that the Southern English of his day (especially towards Devon) was more “incomposita,” and at the same time more archaic than that of the North, he goes on to add: “Originalis linguæ proprietatem et antiquum loquendi modum magis observat. Cujus etiam rei non solum argumentum, sed et certitudinem inde habere potes, quod omnes libros Anglicos Bedæ, Rabani, regis Æluredi vel aliorum quorumlibet, sub hujus idiomatis proprietate scriptos invenies.”⁵⁵

This statement might seem to imply that in Giraldus's day the continuity of English composition from Anglo-Saxon times had never been broken, and this impression is borne out when he says that the English, “like the Welsh,” use alliteration “in omni sermone exquisito” (p. 187). It is here of interest to compare the statement of Mr. Macaulay that “in a certain part of the Southern dialect-region the language of the ‘Katherine-group’ seems to have attained for the time almost to the position of a literary standard” (p. 150). The Katherine-group is written in alliteration, and it may therefore be that this Southern and alliterative literature represents the very same development in English literary history as that of which Giraldus is speaking—and may even make part of the work he had in mind. The fact that the Ancren Riwle, which is no less “elegant” than the Katherine-group, is not written in alliteration, is perhaps due to the original's having been French. Other alliterative works of the Southern dialect region existing in manuscripts of the late twelfth and early thirteenth century are the ecstatic rhapsodies so often connected with the Riwle by similarity of spirit as well as of form.⁵⁶ It is of interest that one of these has been shown to be a translation of a poem by Marbodius of Rheims, who had addressed another poem to “good Queen Maud.”⁵⁷

Whether or no the Middle English composition still extant from the earliest period can be identified with that which is referred to by Giraldus, it is evident that his statements would assist any hypothesis putting English composition in the twelfth century, and some other facts should be noted in this connection. It may be that the extreme conservatism as to inflections, which has been noted as a striking characteristic of the Southern dialect of the earliest period, by comparison with Northern and Midland,⁵⁸ is due in part to the fact that Southern works have been dated too late. Just what Giraldus means when he calls the Southern dialect “incomposita” is not clear, but it is just possible that we should connect with this remark a phenomenon of the Southern dialect which gives a contrast to its syntactical conservatism, namely its abundance of Romance words. This characteristic is most marked in the Ancren Riwle, and for this fact the present hypothesis, by connecting the work with the court circle, where the French influence was strong even before the Conquest, would give a reason.⁵⁹

In general, it is probable that many of the Middle English characteristics of the language of the earliest monuments after the Conquest existed before that time in spoken speech, and only appeared when they did in literature, because the discontinuance of schooling in English had torn away “the veil which literary conservatism had thrown over the changes of the spoken tongue.”⁶⁰ The fact that this earliest Middle English, affected by French orthography, has been thought to show in its phonology the influence of the large element in the population of England after the Conquest who did not know English, would support assigning it to an early date; for, as I have shown in an earlier article,⁶¹ the aliens of the Conquest seem to have been assimilated by the latter part of the twelfth century. The fact that the present hypothesis would put in the twelfth century a large part of the English compositions usually ascribed to the thirteenth need not alter our conclusion; for, as I have shown in the earlier article, though the thirteenth century was a time when English was probably more strongly entrenched in familiar use than ever before, this was also the period of the international use of French in literature, due to the international influence of the University of Paris. This was a time when English composition was much less likely to flourish than in the years immediately following the English revival in the time of Henry I; for, in the latter period, the continuity with Anglo-Saxon culture was not yet lost, and the court was less cosmopolitan than in the days of the Angevins. The very breadth of the Angevin Empire must have been an influence operating to extinguish local usages, especially since the international ideal was animating the intellectual world at that time. It would be natural that revival of composition in English could not be more than a flash in the pan until the international ideal had passed its zenith. The conditions making for the use of a universal vernacular were too easy in England in these years to be neglected when the desire for such an institution was in the air all over Europe.

(XIII)

The Ancren Riwle shows special and even minute reminiscences of the preoccupations of the decade of the foundation of Kilburn. One of the strongest links connecting the Riwle with Kilburn is the relation which that work can be proved to bear to the controversy between the Cistercians and the older Benedictines, which was one of the most outstanding incidents of this time. In this connection the Riwle is always to be found on the side of the older Benedictines, as would be natural to a treatise written for the cell of a Benedictine house.

Vincent of Beauvais, in the Mirror of History which became the standard reference book of the Middle Ages, notes for the period with which we are here concerned the rise of the new orders as a striking characteristic of the time (27, vii). Most of these new orders made settlements in England during the last years of the reign of Henry I, and in the reign of Stephen William of Newburgh tells us that “many more monasteries were built in that short time than in the hundred years previous.” He compares the strength thus manifested by the Heavenly King, with the weakness of the earthly sovereign.⁶²

We have already seen that the enthusiasm of the time for rules and orders was such that, for the sake of a settled connection, many old houses at just this period were assuming the Augustinian rule, and that even the Pope was recommending such action. The eagerness for a rule which the author of the Ancren Riwle tells us that the sisters of the anchorage have shown, is a similar sign of the times, as is also the identical desire which St. Aelred met from his recluse sister. Many of the new orders had developed from hermitages, and we can see the contemporary expectation in the fact that the three ladies of the anchorage are receiving enquiries from others as to “of hwat ordre” they are (p. 8). The author, as we shall see, belonged to a liberal minority in his generation who took a very spiritualized view of orders and of the differences in material usages on which they were built up, and he therefore gives the anchoresses the ironical reference to the “order of St. James,” with which to meet their enquirers. Nevertheless, he proceeds to give—in his own spiritualized form—a rule for the anchorage, though he has denied them an order. Perhaps he does this because he recognizes the very real paucity of rules for women, which is noted by Heloise when, in this very decade, she writes to Abelard for a rule for the Paraclete.⁶³

At the time of the foundation of Kilburn the most aggressive and successful order in Christendom was the Cistercian, which, to use the words of William of Malmesbury (p. 380), then seemed the “surest way to heaven.” It had originated in a secession from the monastery of Molême by monks who had come under condemnation of conscience because the Benedictine rule was no longer observed in minute accuracy.⁶⁴ The essence, therefore, of the Cistercian idea was the meticulous observance of the Rule, and the assumption of superiority that it carried with it aroused the ire of the older Benedictines, who were especially irritated because the Cistercians had changed the black Benedictine habit for a white one.⁶⁵ When in 1122 Peter the Venerable became Abbot of Cluny, a quarrel was raging between his house and the Cistercians because of one Robert (a monk whom both claimed), a relative of Abbot Bernard of Clairvaux, who had written the young man a miraculous letter condemning the laxness of Cluny, which he had dictated—“in medio imbre sine imbre”—to the Englishman William, already mentioned as the first abbot of Rievaux.⁶⁶ Peter, who was the great peacemaker of his generation, entered the controversy with a letter to the Cistercians soon after his accession in 1122. This letter echoes the argument in terms similar to those found in the Ancren Riwle: since the treatise uses a metaphor strikingly employed by Peter, it would appear that his letter was known to the author, at least by hearsay. In any case it gives the best possible setting forth of the older Benedictine side of the controversy, and will be summarized here very briefly on that account.

After a discussion of many particular observances—many of them matters of costume—in which the Cistercians make a great point of differing from the Cluniacs, the abbot sarcastically describes the scorn felt by the black monk for the white, and vice versa, and he condemns the puerility of the procedure. “Corporalis exercitatio ad modicum utilis est” (I Tim. iv, 8; c. 151); it was the very essence of the Rule of St. Benedict to leave matters such as food, drink, clothing, to be decided according to time, place, manners, and other accidents (c. 126). “Hoc tamen reducere ad mentem charitas vestra debet quod divina mandata, partim mobilia, partim sunt immobilia. … De immobilium numero dicimus esse … illud maximum et primum dilectionis Dei mandatum, secundum quoque huic simile de proximi dilectione; sed et humilitatis, castitatis, et veritatis præcepta” (c. 148). He repeats several times over the words of St. Augustine: “Habe charitatem, et fac quidquid vis.” “Haec est quæ plenitudo legis et finis præcepti est (Rom. xiii, 10). … diversis temporibus per diversos sanctos diversa loquens, ipsa non varia, non divisa, non multiplex, sed simplex, stabilis, inconcussa, semper eadem perduravit. Ut enim materfamilias, tota domus propriæ utilitati intenta, quosdam famulorum ad bobus exercendam, quosdam ad fodiendam vineam, alios in silvam ad ligna cædenda mittit … ipsa (charitas) tamen, quamvis diversa sint quæ præcipit, diversa non efficitur, nec diversitas jussionis diversitatem facit utilitatis, quoniam ad unum quiddam et simplex, hoc est ad domum utilitatem, omnis illa officiorum varietas se colligit, et illud unum innumera ministeria operantur. …” (c. 154).

In the introduction to the Riwle the scheme is laid down for the whole work, and it seems to echo the ideal expressed by Peter the Venerable. There are two kinds of rule which the author will describe. The one is the rule of charity, the other that concerning material observances,—all of which may be changed for good reason, provided the rule of charity be preserved. Charity is the “Lady Rule” who makes “all things smooth”—to whom the other is only the “hand-maiden.” This general scheme is again mentioned just before the summary (p. 410). The three crucial texts used by Peter above quoted, appear prominently in the Riwle (pp. 4, 386). The appearance (pp. 2-6) of some sentences in Latin would seem to show that the author is using some definite source, which we may hope to trace when manuscripts can be investigated. …

The account already given of Peter the Venerable's letter will have shown the significance of the discussion in the Riwle of “black” and “white.” Other reminiscences of the differences between the two orders appear in the passage just quoted. The phrase “strain at a gnat and swallow a camel” was used commonly in this controversy. Peter applies it (c. 124) to the Cistercians in his letter already quoted, and he says elsewhere of his beloved Carthusians that they do not “strain at the gnat,” etc. (Opera, c. 412). Bernard of Clairvaux, on the other side, applies it to the Cluniacs in his Apologia.⁶⁷ In the same epistle (c. 913) Bernard mentions (to refute) what are almost the words of the Ancren Riwle: “Cæterum in habitu, inquis, non est religio, sed in corde. Bene. At tu quando cucullam empturus lustras urbes, etc.” However, his epistle to Robert will show how naturally his generation, as it seems, chose this saying for meeting his arguments. In this epistle he attacks many petty details of the life at Cluny, as the following quotation will show (the italics are mine and mark details referred to in the Riwle):

“Si pelliciae lenes et calidae, si panni subtiles et pretiosi, si longae manicae et amplum caputium, si opertorium silvestre et molle stamineum sanctum faciunt; quid moror et ego quod te non sequor? Sed haec infirmantium sunt fomenta, non arma pugnantium.

(c. 77)

“Frocci” (generously made as to materials, with wide hoods, and sleeves),⁶⁸ furs and “stamins,” were some of the luxuries which the Cistercians discarded, but the anchoress is told that her sin is the same, “however her kirtle be shaped or colored” (p. 200). The argument that raged during this generation as to the shape, size, material, and color of monastic garments certainly echoed in every phase of the declaration of the Riwle that “religion is not in the kirtle or the cowl, or in the wide hood, or in the white, black or grey cowl.” The “grey cowl” is probably a reference to the order of Savigny: their founder and his monks had been received into the fraternity of Westminster, and their greatest house, Furness, had been founded by Stephen of Blois in 1126. The Middle English “Winteney Rule” (Cistercian) rejects “cowls” for “kirtles.”⁶⁹

Peter the Venerable, in the letter already quoted, gives us a hint that the Cistercian quarrel was with the size of garments, as well as with the color (grossitudo et color). But he proves that the Cistercians are here the innovators: St. Martin did not go “in albo et curto (the italics are mine), sed quod nigro et pendulo pallio” (c. 116).⁷⁰ He discusses at length the inhumanity of the Cistercians in refusing to allow their monks to wear furs (cc. 120 ff.), and his statutes show that his monks wore stamins after their reformation (c. 1043). The anchoresses (like St. Aelred's sister, c. 1458, who here received a grudging concession) are allowed both stamins and furs and (unlike the Cistercians) given as many garments as they need (p. 418). …

Here we have again the wide hood, and sleeves which might from their width appear “long,” and the older Benedictine costume may be in question: the Cluniacs, as we have seen, according to Peter the Venerable do not appear to have developed novelties in their costume. At Westminster, according to the Customary (pp. 140, 146 f.), frocks, cowls, stamins, and furs were all worn. The “close cope” was required of all religious persons. It is very possible, therefore, that the author is here in a playful manner elsewhere used directly referring to the monks of Westminster, who, as we know, at a later date probably made a great deal of trouble by trying to see the nuns of Kilburn.⁷¹

The Cistercians have been described as the “Puritans of the Middle Ages,”⁷² and their controversy with Peter bears out this title. He, on the other hand, seems to have found the finest “Via Media” of his generation. It would appear that his eloquence must have made him the spokesman for the “black monks” in general; for the older usages scorned by the Cistercians seem to be those of all older Benedictines, who were all “black monks.” The division between Cluniac and other Benedictine houses was so purely a political one, that in England, at any rate, Cluniac monks were constantly presiding over Benedictine houses,⁷³ and when Henry I made his great foundation at Reading, he settled it with Cluniac monks, though the house was counted Benedictine.⁷⁴ It is of great interest to the present hypothesis that Peter the Venerable visited England in 1130,⁷⁵ and, considering that the difference in political organization did not separate the English Benedictines from Cluny, we may feel certain that he came to Westminster, and that discussion of the great differences of usage and of spirit between the “black” and “white” monks took place during his stay there. The monks of Westminster would be able to sympathize with the stand he had taken in the controversy, because the distinction between the “movable and immovable” ordinances, on which it was founded, was to be found in brief but complete form in the statutes which had been drawn up by Lanfranc for the monks of Canterbury,⁷⁶ and this work we find used later in the Westminster Customary.⁷⁷ Since Peter took an intense interest in the nunnery of Marcigny (which included anchoresses), there may have been discussion of plans for Kilburn.⁷⁸

The events which immediately followed Peter's visit were such as to have enhanced the interest of his defence of the “black monks.” In 1132 he reformed his order, somewhat along the lines of St. Bernard's criticism, though the bitterness between the two orders was not assuaged in consequence, and in his famous second letter on the controversy, written in 1143, the “black” and “white” monks are still described as turning the other way when they met.⁷⁹ In England in 1132, as we have seen, some monks seceded from a Benedictine house to found Fountains Abbey, and the enormous development of the Cistercian power in England had begun. Henceforth, for the next two decades, the material success of the Cistercians in England must have kept their controversy with the older Benedictines very much alive.

Two landmarks of the literature that the controversy must have produced in England can be pointed out and grouped with the Ancren Riwle, and doubtless others can be found in manuscripts. One of these is the work of St. Aelred, one of the greatest English Cistercians. He had been brought up in the family of the brother of “good Queen Maud” (who lived at the English court till 1124), and he was called in by the abbot of Westminster to revise for the canonization the life of Edward the Confessor (which had been prepared by Osbert of Clare),⁸⁰ as he had been called in by Gilbert of Sempringham to advise as to the affairs of the Gilbertine order. Aelred was a central figure in the life of his time, and it is therefore not surprising that just as his letter for recluses⁸¹ parallels both the Riwle and an epistle of Peter's, so now another work parallels both our treatise and the epistle of the abbot now under consideration. St. Aelred Speculum Charitatis was written at Rievaux (under the rule of William, the scribe of the epistle of St. Bernard to Robert), between the years 1140 and 1142-3,⁸² and he devotes one chapter of this work to combating a letter “cujusdam” which has declared that the ultimate essential which makes a monk (c. 608) is only the three-fold vow. “Quid mihi, inquit, de Regula objicis? Habe charitatem, et fac quicquid vis” (c. 612). The quotation, “Habe charitatem” (used in the Riwle and three times by Peter in his letter already quoted) is not the only locus communis which this work shares with the epistle, but the letter which is here combated (and quoted verbatim) is not Peter's.

The other similar echo of the spirit of the times—it should probably be given no more explicit title—occurs in the great Polycraticus of John of Salisbury, written during his residence (1150-61) at Canterbury. In his chapter on hypocrites (Lib. VII, cap. xxiii) he declares “Regula namque verissima est, quod de radice caritatis non oriuntur mala, et de cupiditatis planta nequaquam bona provenient.”⁸³ He recognizes the virtue that can be found in every variety of religion, even in his own estate as a secular clerk, and he quotes St. Jerome (p. 184), “vestis nequaquam religionis differentiam faciat.” His conclusion is that of the author of the Riwle: “Novi tamen regulam veritatis, qua mihi constat, quia religio munda et immaculata apud Deum et patrem, haec est, visitare pupillos et viduas in tribulatione eorum, et immaculatum se custodire ab hoc seculo. Haec autem politicorum omnium est: et bene cum istis agitur, si eam fideliter servant” (p. 185). It is evident that the “order of St. James” of the middle of the twelfth century numbered others besides the anchoresses. Peter of Blois (Archdeacon of Bath), the pupil of John of Salisbury, should also be included, for he also repeats the text from St. James in the same context, preceding it by the saying: “regnum Dei intra vos est, non in exteriori vestitu.”⁸⁴ Peter of Blois carried his intolerance of the forms of religion so far that he would not even take priests' orders.⁸⁵

John of Salisbury and Peter of Blois, secular clerks living at courts, and Peter the Venerable, abbot of the most sumptuous order in Christendom, may appear unexpected spiritual relatives for the anchoresses of the Ancren Riwle; but the relationship does not appear so surprising when we observe that John and his pupil,⁸⁶ like the abbot, express an extreme admiration for the Carthusians, who follow a mode of life very similar in ideal to that described in the Riwle (which can be proved to have been known to the author of that treatise). We have again evidence that the present hypothesis puts the Ancren Riwle in the right environment. At the middle of the twelfth century—in religion, philosophy, and the arts, one of the most creative periods in history—the liberal party of the church evidently combined an unusual elasticity as to the forms of religious life with a most intense concern to avoid hypocrisy (which also appears in the Riwle, v. supra, p. 522). Consequently the forms of monasticism that gained its special affection were those, like the life of a Carthusian or of an anchorite, in which, according to the phrase of the time, the members lived “in a sepulchre.” This somewhat unusual combination of ideals appears in the Riwle. To understand the full contemporary meaning of that work, we must read the letters of Peter the Venerable; and we must read the Ancren Riwle to understand the spirit that actuated Abbot Herebert to bestow the goods of his abbey to perpetuity, on persons of the other sex who did not even come under the title of his own order. The sectarianism common to all ages was for a moment in abeyance, for the unspiritual importance attached to differences in usage in that age of the rise of many orders had stimulated the liberals of that generation to special magnanimity.

Some details may be briefly mentioned which might serve as obstacles to the early date here given to the Riwle.

The most conspicuous obstacle comes from the fact that St. Bernard (either with or without the “saint”) is quoted thirteen times. As has been stated above (p. 529 n.) citations cannot certainly be taken as original: in any case the author was a contemporary of Bernard, and the names may have been added later, during the circulation of the work for nearly a century. An anonymous work would inevitably gather such accretions. Quotations from early writers are in any case greatly in the majority.

An obstacle may seem to exist in the apparent references (Morton, pp. 16, 32, 268) to the elevation at mass, since this has been said to have come into England at the beginning of the thirteenth century. Father Thurston, however, in his epoch-making analysis of the development of the later elevation from an earlier (always made), declares that the passages in the Riwle do not give certain information whether the earlier or later elevation is referred to,⁸⁷ and one seems to imply the earlier.

Another liturgical practise found in the Riwle that may appear to be of late introduction is the hanging of the Sacrament over the altar (p. 16). This, however, can be carried back to the time of the Riwle, because the breaking of the chain which suspended the pyx over the altar was one of the ill omens at King Stephen's mass at Lincoln in 1140.⁸⁸

Still another usage found in the Riwle that has been said to belong to the early thirteenth century is the frequent use in private devotions of the Ave Maria,⁸⁹ but later research has shown that among ascetics this custom goes back very far.⁹⁰

The present paper can be only a preliminary statement of the theory which is here suggested, and the general historical relations which make it so interesting cannot be more than touched on here. Since, however, in another article⁹¹ I have hinted at some of them, I must briefly return to the subject.

We have evidence from the historical side of a religious revival in England during the reign of Stephen, of which the spirit expressed in the mystical English works earlier grouped together (supra, p. 513) is exactly characteristic. Passages quoted in my earlier article in which St. Aelred describes the devotion of Gilbertine nuns should be put side by side with the “Katherine group,” or the ecstatic rhapsodies, for example, to show how likely it is that these pieces—also strongly mystical and specially concerned with the devotion of women—should have emanated from the same environment.⁹² The most actively germinating element in the England of Stephen was monasticism, and it would be very likely that it expressed itself in literature. Since, as we have seen, the Southern dialect at this time was used as a literary standard, it may be that some of the English mystical works written in Southern speech originated in the Gilbertine houses of the North Midlands.⁹³ Perhaps they were written by Gilbert himself, who, as we are told by his contemporary biographer, “wrote books” (p. x). From the historical side, therefore, the reign of Stephen would be a most likely time to put the mystical works in question, and the mystical rhapsodies in any case bear a very close relation to our treatise.⁹⁴ The “Katherine group,” like the Ancren Riwle, shows a mixture of late and early forms.⁹⁵

It must never be forgotten that the chronicle of Gaimar, which was written in the decade here suggested for the composition of the Ancren Riwle,⁹⁶ is preserved in only one manuscript which is as early as several in which the Riwle is found. The De Nugis Curialium of Walter Map exists in a unique manuscript written two hundred years after the composition of the work. There is therefore no reason why the Ancren Riwle cannot be put back to the reign of Stephen, in spite of the lack of manuscripts of that period, if the evidence to do so is sufficiently strong. It is the purpose of the present paper to show an accumulation of coincidences that will localize the time of composition of the work as strongly as would the mention in the text of an author whose date is beyond question. I believe that the connection of the Ancren Riwle with Kilburn is supported by a range of evidence surprisingly exact and complete, when the remote period of history is considered with which we are dealing. The treatise was composed in the most important circles of its day, and we find our problem of tracing its connections made easy for us in consequence. The history and connections of Kilburn can be followed during a long period, and they agree so strikingly with the history and influences indicated for the anchorage of the Riwle, that it would seem impossible that two establishments could present such a series of parallels, sustained throughout such a term of years. Many questions connected with the history of the Riwle have been touched on in the preceding study, but they are all subordinate to the identification of the anchorage of the treatise with the hermitage of Kilburn. That identification stands and falls solely on the strength and peculiarity of the parallels drawn from the specific statements of the treatise, on the one hand, and from the historical records of Kilburn on the other. No conjectural evidence whatever enters the discussion at this point, which is the cardinal one of the whole investigation.

APPENDIX

The present section will comment on the article by the Rev. Vincent MacNabb, O. P., in the Modern Language Review, xi, pp. 1 ff., which seeks to ascribe the Ancren Riwle to the authorship of Robert Bacon, O. P. The earlier claims to the authorship made for Simon de Ghent and for Richard Poore have been conclusively refuted, as Mr. Macaulay makes clear (pp. 77-8): Simon de Ghent is apparently the author of the Latin version, and the connection with Tarente, on which the conjecture as to Poore's authorship is entirely founded, belongs only to this form of the work.⁹⁷ Many of Father MacNabb's arguments have already been either matched or refuted, but others require some special mention.

The basis of Father MacNabb's whole article is the two passages (Morton, pp. 24, 412) in which the anchoresses are given the usages of “our lay-brothers.” Since the directions given almost—not entirely—coincide with Dominican practise, Father MacNabb concludes that the author was a Dominican. His hypothesis therefore loses its foundation, when we discover that the passages in question must certainly be interpolations, perhaps due to a Dominican scribe.⁹⁸

The reference to the lay brothers on p. 24, in which the anchoresses are told to say their hours by Pater Nosters, like these illiterate persons, is absurdly out of keeping with the rest of the work; for the sisters have already been told (p. 20) to “say their hours as they have been written,” and it is very evident that they are able to read both Latin and the vernacular. If it be urged that the directions in question are intended for the serving maidens, it can be answered that they are then superfluous and contradictory; for these have been already provided for in the hours to be used by the sick and the illiterate. … The number of Pater Nosters here arranged for differs entirely from those of “our lay-brethren,” and they, rather than the latter, seem referred to for the lay-sisters in the concluding summary; “Si ele ne siet rien deliure, die par pater nostres et par aueez ses houres” (f. 69, Morton, p. 424); the latter contain no Aves, whereas these bear an important part in the former.

A further fact lessens the authority of the passage. It is found only in N, and though, as we have seen, this copy at p. 192 gives a unique text which we believe to be the original, the two cases differ entirely. The authority of the earlier divergence is upheld by the partial agreement of other manuscripts, and by evidence found in other parts of the work, whereas in the present instance the omission of the passage from all other copies is the least suspicious detail in its connection.

The passage, it should be noted, gains in continuity by the omission of the reference to the lay-brothers. The sentence, “In this manner you may say, if you will, your Pater Nosters” then applies to the elaborate series of Pater Nosters which follows—three for the Trinity, five for the Five Wounds, etc.—and this complicated devotion is a way of “saying your Pater Nosters” very suitable to the cultivated anchoresses. The French text seems to make this application certain, for it makes of the sentence a heading parallel to many in this text: “Coment vous deuez dire voz pater nostres en lonourance de la trinitee. Dieu tout puissant,” etc. (f. 5v).

The second reference to the lay-brothers can be as decisively disposed of as the other on the basis of the French text, which here omits the reference (though Mr. Macaulay has neglected to note the fact) as follows: “A queux jours vous deuez estre accommunez (rubric). Len tient le miens de la chose qe len ad souent, pur ceo ne deuez vous estre (the English here introduces “bute ase ure leawude breðren beoð,” p. 412) acuminez dedenz lan fors quinze foiz” (f. 67). In the ensuing list of dates the more precise “Ascension Day,” “Christmas,” and “le jour seint Johan le Baptiste” are substituted for the “Holy Thursday,” “Midwinter Day,” and “Midsummer” of the English copies.⁹⁹ All the circumstances would make it appear that the French here offers the original text, and its virtue is again proved.

The “connection with Salisbury diocese” brought by Father MacNabb into his argument (p. 5) belongs, as has already been noted, only to the Latin text. It should be pointed out, however, that the present hypothesis, by connecting the work with Westminster, offers a perfect explanation for the interest shown in the work by the anchoresses of Tarente, especially if attached to the nunnery. This house was taken under the special protection of Henry III, the rebuilder of Westminster; the king's sister was buried there, and Richard Poore bequeathed his interest in Tarente to Queen Eleanor. It would be very likely, considering the special interest of the king in both abbeys in the time of Simon de Ghent, that the women at Tarente would learn of the treatise written for the only house of women connected with Westminster. It may be that we have here a case of the propagation of the Riwle, like those referred to in B.¹⁰⁰

Father MacNabb refers (p. 6) to the Mirror of St. Edmund as written for women, but it is addressed in the first line to men, and was written for the monks of Pontigny.¹⁰¹

Some of the points made by Father MacNabb can be turned to the account of the present hypothesis. This is true of his reference (p. 5) to the mention of the house addressed in the additions of B, as “the cloister over which Jesus Christ is high Prior”—of interest because Kilburn was by this time called a Priory.¹⁰² His finding support for his theory (p. 4) in the mention of friars by B is surely a boomerang, for if the work were written by a friar, or in the time of friars, it is very strange that they are not once mentioned in the original text.

A group of details connected by Father MacNabb with Dominican influence can be shown to be of general use in the church. He refers to “venia” (p. 4) as a “technical phrase used daily by Dominicans,” but a reference to Ducange will show that it was in constant use by all religious orders. In the blood-letting four times a year (p. 3) also, the Dominicans were not peculiar, since the Cistercians did the same: moreover both these orders set seasons, whereas the Riwle, the Benedictines and the Cluniacs did not.¹⁰³ The blessing of a drink taken between meals (ibid.) would seem too natural to need mention, were it not that we are told in connection with Wulfstan of Worcester, the last of the Old English bishops (who always practised it), that it was specially characteristic of the English.¹⁰⁴ His chaplain (on the death of the bishop in 1095) became a monk at Westminster, where he lived for twenty years—as we are told by Osbert of Clare.¹⁰⁵

The directions to the sisters to sit and stand alternately during psalmody, to rise and bow at the Gloria Patri, and to sit at Placebo till the Magnificat (MacNabb, p. 3) find precedent in the customs of Augustinian canons, who divided their convent into two choirs, rising and sitting alternately, and bowing “choir to choir” at Gloria Patri. They sat at Placebo, except at certain psalms, of which the first was Magnificat.¹⁰⁶ It is significant that the question of saving the religious person from over-great fatigue from long-standing was one much under consideration at the time of the composition of the Riwle.¹⁰⁷

In conclusion, Father MacNabb compares (p. 6) the passages in the Riwle describing the austerities practised by a friend of the author, with similar ascetic practises attributed to Edmund Rich by Robert Bacon. This analogy can be shown to be neither exact nor peculiar.

The use of austerities like those in question was extremely common among the more ascetic persons of the twelfth century, as reference to Miss Clay's Hermits and Anchorites will show, and earlier they had made part of the special message of Peter Damian. An example can be quoted from approximately the same time and environment as the first ladies of Kilburn, which fits the description of the Riwle more closely than that given by Father MacNabb.

William de Lacie, the hermit who, with Ernisius chaplain of “good Queen Maud,” founded Lanthony Priory, wore articles which may be quoted side by side with those mentioned in the Riwle, and those ascribed to St. Edmund¹⁰⁸. …

Since the lorica had sleeves, the binding of Lacie's costume may duplicate all the “bands” of the Riwle. It should also be noted that he wears all these instruments continuously, as also seems to be true of the person described in the Riwle, whereas St. Edmund wears some of his only during Lent and Advent.¹⁰⁹ Since the person described in the Riwle also “fasts, wakes and labors,” it is natural to suppose that he must be someone quite free from the occupations of the active life.

It would appear that the theory of Bacon's authorship can be refuted as conclusively as the two earlier on the field. When we reject the references to the lay-brothers, it loses its basis: nevertheless, Father MacNabb, by pointing out significant details in the Riwle hitherto unnoticed, has made a permanent contribution to the study of the treatise.

Notes

1. All the manuscripts and versions of the work are carefully described in the articles by Mr. G. C. Macaulay in the Modern Language Review, Vol. ix, passim. It is Mr. Macaulay's opinion that the French version is the original, and that the Latin version is a late translation. The nomenclature which he uses for the various manuscripts will be reproduced here. The earliest manuscript, B (of about 1230, twenty years earlier than N, G, T, and C), is the most correct, but also the most interpolated. He prints collations of all the thirteenth-century copies, and the complete text of the nineteen passages added in B, some of which are reproduced in later manuscripts. N is the only copy in print (ed. Morton, Camden Society, 1852). I possess a rotograph of parts of the unique French text—an early fourteenth-century manuscript.—G gives extracts only.

2. Monasticon Anglicanum, London, 1821, iii, pp. 422 ff. By the kindness of my friend Miss Deanesly of Newnham College, I possess a copy of the same charter from the Westminster Liber Niger Quaternus, f. 125—a collection of documents made at the abbey 1466-85, from one of 1408-11 (see The MSS. of Westminster Abbey, by J. A. Robinson and M. R. James, Cambridge, 1909).

3. The grant is made “in the time of Henry I” by Abbot Herebert and Prior Osbert de Clare, with the consent of Gilbert the Universal, Bishop of London. The latter was bishop between 1127 and 1134, as is generally supposed, though Wright notes that some authorities put Gilbert's death at 1138 or 1139 (Biographia Literaria, Anglo-Norman period, London, 1846, p. 103). It is generally accepted as having taken place on the way to Rome, August 10, 1134, as is recorded in the necrology of Auxerre, where he had been head of the schools before coming to London (see Histoire littéraire de la France, xi, p. 236). The confusion perhaps arose because after his death there was a vacancy in his see till 1141. The dates formerly given for Herebert's abbacy were 1121-40, but Dr. Robinson has lately shown that his successor, Gervase (natural son of King Stephen, who was crowned December 26, 1135) was “already abbot in 1138, if not sooner” (History of Westminster Abbey by John Flete, Notes and Documents relating to Westminster Abbey, No. 2, Cambridge, 1909, p. 142). The letters of Osbert of Clare show that he was exiled before 1123, and was “proscriptus” in 1133. He may have been back during 1127-28, and was back as Prior in 1134. Dr. Robinson puts the foundation of Kilburn in 1134, but it seems difficult to be certain on this point, considering that Gilbert the Universal, if he died abroad August, 1134 (as Dr. Robinson believes) must have left England early in the year, and he granted a second charter in connection with Kilburn—“Cum inter nos et abbatem Westmonasteriensem et conventum quaestio verteretur super subjectione et jurisdictione Cellae de Kilebourne” (No. iii in Dugdale). See Dr. Robinson's article on Osbert (Church Quarterly Review, July, 1909, pp. 437-54), and the latest volume in the Notes (The Monks of Westminster, by the Archdeacon, E. H. Pearce, Cambridge, 1916, p. 42). Some of Osbert's letters, in abridged texts, have been printed by R. Anstruther, Caxton Society, 1846. Among those unedited may be some of great interest for the present hypothesis, for Miss M. Bateson notes that Osbert was a correspondent of the Kilburn nuns (Mediaeval England, London, 1904, p. 91—the reference which first brought Kilburn to my notice).

4. The most complete account of its history will be found in J. J. Park's History of Hampstead, London, 1814. An article in the Transactions of the Hampstead Antiquarian and Historical Society, 1904-5, pp. 90-101, contributes nothing. The site of the Priory has now been absorbed in London, but a hundred years ago it was still to be distinguished, and two hundred years ago there were still fragments to be seen of the building. Some trifling relics have been excavated on the site, of which the most interesting is a brass showing the head of a prioress. The wimple is said to be unique (Proceedings of the Society of Antiquaries of London, 1885-7, 2nd Series, vol. xi, p. 23). Until a late date the locality seems to have been distinctly rustic; it was five miles from London (London and Middlesex Archœological Society, 1883, pp. 273-7). Its literary associations have been famous: Goldsmith perhaps wrote some works here, and Keats first recited to a friend the Ode to a Nightingale in the “Kilburn meadows.”

5. The anchoresses are addressed as young often (Morton, p. 56, etc.); as three in a passage never lacking (Morton, p. 116, see Macaulay, p. 157, Fr. f. 22v); and they are spoken of as of noble birth in another passage always present (Morton, p. 356, Macaulay, p. 328, Fr. f. 57v). The extended description of their circumstances given by N (Morton, p. 192), which repeats these details, falls in a gap in the French text, and is not entirely reproduced in any other copy (Macaulay, p. 159). B, as we shall see, had good reason for omitting it, and has only by inadvertence preserved the personal references noted above; T gives the information as to the stir caused by the maidens' retirement, as it is given in N, and adds the details as to their high birth and youth in other language that seems chosen for the sake of alliteration (this copy has, in general, Northern peculiarities). The general statement as to the material ease of the sisters' circumstances, with which N begins this passage, is repeated by C. The fact that so much of the information given by N in this important paragraph is confirmed from one source or another, seems to show that N is here giving the original text, and to guarantee the accuracy of one further piece of information here given, for which N is our only authority (see infra, p. 482).

6. Op. cit., p. 87. Dr. Robinson is inclined to doubt Flete's information, principally because of the lack of provision for prayers for Queen Matilda in the foundation (C. Q. R., p. 353). Her anniversary was however kept at Westminster (see the Customary of the abbey—compiled about 1266—printed by the Henry Bradshaw Society in company with the related Customary of St. Augustine's, Canterbury, 1904, p. xxvii); and the nuns of the Ancren Riwle, as we shall see, kept the anniversaries of their patrons. It should be noted that Matilda died in 1118, and Kilburn, according to Dr. Robinson's dating, was not founded for sixteen years. Some of her “domicellae,” however, might at that time still be “puellae,” since they might have served her as children, according to the custom of sending children to be reared at court. She lived, as it were, across the way from the abbey for many years at the end of her life (William of Malmesbury, Gesta Regum, Rolls Series, 1889, ii, p. 494), and the maidens may have come to the attention of Abbot Herebert when at the palace. Widmore has already seen reason to believe that he was in close connection with the court (History of the Church of St. Peter, Westminster, London, 1751, p. 22). Dr. Robinson notes that Osbert of Clare (whom he considers to have been the moving spirit in the foundation of Kilburn) was “in 1121 of sufficient importance for the king to interfere with his plans” (C. Q. R., p. 339). The Ancren Riwle does not necessarily imply extreme youth for the anchoresses. N uses the terms “in the blossom of your youth,” and T the words “young of years.” A medieval commentary describes the age of Christ at the Resurrection as “aetas juvenalis” (C. Horstmann, Yorkshire Writers, London, 1896, ii, p. xix, n.). Persons younger than twenty were not generally admitted to monastic vows (for Cluny, see Migne, Patrologiae Cursus Latinae, clxxxix, c. 1036).

7. It may be that the term “inclusa” was sometimes applied to any nuns who were bound by vow not to leave their house, whether they were confined in solitude—more or less complete—such as is the case in the anchorage of the Riwle or not. It is hard to tell the meaning of the term as applied to the first nuns of Sopwell Priory, for example (a cell of St. Alban's—see De Gestis Abbatum, Rolls Series, 1867, i, pp. 80-2). About 1140 two women who had lived in a wood were given a more regular establishment by the Abbot of St. Alban's, who “added others, and gave them the Rule of St. Benedict,” as Sopwell Priory. The number “inclusarum” was not to exceed thirteen, and “easdem sub clave et sera, ac sigillo Abbatis, qui pro tempore erit, decrevit ibidem includendas.” This ceremony suggests the life of anchorites who, according to the Rule of Anchorites by Grimlaic (Migne, ciii, cc. 573 ff.)—a work certainly influential for the Riwle—were to be formally sealed in their cells. However, no further information is forthcoming from any quarter which would show the Sopwell nuns to have been dedicated to solitude. In 1338 they were not solitaries (De Gestis, ii, pp. 511 f.). The writer in the Victoria County History (Hertfordshire, iv, pp. 422 f.) believes the nunnery to have been founded in recompense for removing the women from the monastery (St. Alban's had been a double house), and he also notes a confusion in the records between Sopwell and Markyate. The latter Priory was dedicated in 1145 as a cell of St. Paul's, but it had been in close connection with St. Alban's. It was founded by “St. Christina of Markyate,” perhaps the most famous anchoress of her time, who had lived in solitude under the protection of a hermit Roger, a monk of St. Alban's (see De Gestis, i, pp. 95 ff.). It is certain that in later times Markyate was not a house of enclosed nuns, from incidents when the bishop visited it to explain the statute of Boniface VIII, “De Claustura Monialium” (V. C. H., Bedfordshire, i, p. 359.—The Cluniac nunnery of Marcigny was at this time famous throughout the world, and one of its most noteworthy features was the fact that all its inmates had made the unusually strict vow of never crossing the threshold of the house. Some of them were vowed to a stricter life in cells as “anchoretae,” and it appears to be only these who are called “inclusae,” (see M. Marnier, Bibliotheca Cluniacensis, Paris, 1614, pp. 455 ff., 491 ff.; J. H. Pignot, Histoire de l'ordre de Cluny, Paris, 1868, ii, pp. 31 ff.). “Inclusa” is also the term used throughout the letter of St. Aelred to his sister, which has always been taken as describing a life similar to that described in the Riwle (Migne, xxxii, cc. 1451 ff.).

8. The case of well known anchorites who were enclosed in parish churches may have created a preconception in many readers of the Riwle that the same was true of the women there described. As a matter of fact the “church” of the Riwle may easily be that belonging exclusively to the sisters' own establishment. We have seen that the “heremitorium” of Kilburn is called an “ecclesia,” and the same term is applied to the monastery of Westminster itself in the foundation charter of Kilburn. Both for Kilburn and the abbey, the term that is applied to the personnel is “conventus.” It would appear to be a sign of exclusive possession by the anchoresses of the church in which they are enclosed, that they seem to have chosen the dedication of their altars (Morton, p. 18).—The Gilbertine order took its rise at about the time of the establishment of Kilburn in a “church anchorage” of much the same type as that described in the Riwle and in the records of Kilburn. Gilbert of Sempringham enclosed the first seven Gilbertine nuns (village maidens), not then expecting they would again come out, or that more would join them (Dugdale, vii, p. xix). This he tells us himself, in the Gilbertine Rule, and his contemporary biographer writes: “Hoc modo constructis rite domibus religioni competentibus, et claustro circumquaque clauso, inclusit ancillas Christi solitarie victuras, sub pariete ecclesiae beati Andreae apostoli in vico de Sempringham, ad aquilonalem partem … fenestra tamen patente, per quam necessaria intromitterentur” (ibid., p. vii). Gilbert carried the key of their enclosure always with him. This sounds as if these women were enclosed in a parish church (two of which were in Gilbert's patronage, as Lord of the manor), and as if they were to live in solitude. The presence of a cloister need not presuppose elaborate monastic buildings; the famous contemporary hermit, Godric of Finchale, built a cloister in his tiny hermitage (see his Life and Miracles, written by Reginald of Durham at the request of St. Aelred of Rievaux—therefore before 1166—Surtees Society, 1845, p. 153).

9. It is obvious that the nuns of the Riwle enjoy considerable community life, and such is provided for in the Rule of Grimlaic, already mentioned. He describes anchorites living under the protection of a monastery, and he prefers that several should be living together. Peter Damian, in describing the life of hermits, says that if two are living together one is to be the superior (Migne, cxlv, c. 350), and it may be (especially considering the fact that they were living under the protection of a person of the other sex, as we shall see) that one of the three women put in Kilburn was the superior; such an arrangement would appear necessary for executive and legal purposes—in any case if the number in the hermitage were increased. The Carthusians (who, a half century before, had developed what was to be till the end of the middle ages the most successful plan for pursuing the eremetical life in community) lived under priors, as did the group of solitaries among whom lived (in subjection to Cluny) the anchorite Gilbert, to whom Peter the Venerable, Abbot of Cluny, addressed his epistle on the solitary life (Migne, clxxxix, cc. 40, 89 ff., 233 ff., 360 ff.). Peter and his friends the Carthusians will be seen later to have a connection with the Riwle. Some communities of hermits in England are listed by Miss R. M. Clay in her Hermits and Anchorites of England, “The Antiquary's Books,” London, 1914, passim, but there is no record that any were perpetuated. Cardinal Gasquet notes that a group of anchorites lived at Shrewsbury (The Nun's Rule, “The King's Classics,” London, 1907, p. xvi), and the records of their existence there seem to cover a considerable period (see The History of Shrewsbury, by Owen and Blakeway, London, 1825, i, p. 315 n., ii, p. 475). He also notes the Carthusian life as an analogy for that described in the treatise.

10. See infra, p. 492. However the service for the dead seems to have been said by most religious houses daily—even by the Cistercians (see William of Malmesbury, ii, p. 383).

11. The presence in the French text of the reference to the “anniuersiaires de voz chiers amys” shows that a connection with specific benefactors may be intended by the reference to the anchoresses' living by alms, which is less specific than the English, and in part defective: “Ieo apeele honte touz iours estre tenu en despit et mendier sicome vn harlot, si mestier est, sa vitaille, et ses viures daltre almoigne, sicome vous … (defective for perhaps a line at the top of the page) … ent dangier dascune (?) ascu … foiz qi porreit estre votre serf …” (f. 57 f.). …

12. The first statement is found in N and C, the second is found only in N (see supra, p. 476 n.). As is explained earlier, the rest of the information supplied by N in this paragraph is so well substantiated, that there seems every reason to believe that, for the whole description, N furnishes the original reading.—It should be mentioned that the admonition against “gathering” (p. 286) was a stock feature in the contemporary rules for anchorites. It occurs in St. Aelred's letter to his sister (c. 1452) and also in Peter's letter to Gilbert (c. 89 ff.).—One of the conspicuous characteristics of the Carthusians, as described by their friend Peter (c. 994), was their refusal to accept more than enough for their bare sustenance, and Gilbert of Sempringham, at the beginning of his order, was reluctant to receive endowments (Dugdale, p. viii).

13. The three corrodies appear in the charter of Gervase, who also grants a charter confirming the possession by the nuns of the land in Knightsbridge granted by Herebert (“sicut eas inde saysivit frater noster Osbertus de Clara”—a clause that gives Dr. Robinson one of his reasons for assigning to Osbert a principal interest in Kilburn). No sign appears in these charters of Gervase's of the sum of money granted earlier or of the land in Southwark, and it is difficult in all respects to know whether Gervase has altered the endowment of the house. He gives no sign that his charter relating to Knightsbridge is a confirmation, and it may be that his other charter also merely confirms what had been customary, for the foundation charter is sufficiently vague in its reference to the “duo beneficia” to make it possible that three corrodies were really covered by the phrase. Flete says that Abbot Herebert gave the women of Kilburn “certas terras, annonas, et corrodia quaedem cum redditibus de monasterio” (p. 87). It would be natural that three corrodies should be given to three persons, and as it stands the charter seems to divide the spiritual benefits of the first “beneficium” between Westminster and Fécamp. All memory of Fécamp seems lost from the charter of Gervase, but the corrody of Ailmar appears as the first in the list. The last is one “which by their charter they had before,” the second “the corrody of Abbot Gilbert.”—From the latter mention Dr. Robinson believes that the house of Kilburn was planned in Crispin's time (Gilbert Crispin, Notes, No. 3, Cambridge, 1911, p. 34): Archdeacon Pearce (p. 41) is misled by it to believe that the corrody was given by Gilbert (who died about 1118).—The twelfth-century hall of the Abbot of Westminster still exists (The Abbot's House at Westminster, by J. A. Robinson, Notes, No. 4, Cambridge, 1911, p. 4).

14. Monastic houses sometimes made definite provision for the food of servants of pensioners. For example, Vaudey Abbey at its foundation in 1147 was to give corrodies to the founder and his wife, as well as food to their two servants as for two servants of the house (see the account by Miss A. M. Cooke of the settlement of the Cistercians in England, English Historical Review, viii, p. 664). Whalley Abbey in the latter fourteenth century gave to an anchoress and her two servants, bread, beer, and a weekly sum of money, as well as faggots, etc. (see the History of Whalley, by T. D. Whitaker, Blackburn, 1800, i, p. 60). It would certainly seem that the endowment of Kilburn would have provided for all the wants of the hermitage, if the numbers were not increased, and the silence of the documents would make it appear that they had not been.

15. The French here has “voz plus chiers amis” (f. 68), which may recall the terms used in connection with the anniversaries (supra, p. 481, n.). The recension to be described later, which was apparently written for Kilburn, uses the indefinite “good men and women” (Macaulay, p. 471). This is the text found in B.

16. According to the charter of Gervase they received bread, beer, “coquina,” wine, mead, and “pittances,” “cum uno tantummodo clareto, sc. pro anima Gilberti abbatis.” The Customary states that they received four gallons of wine on festivals and anniversaries when the convent received it, and beer also when the convent had it; four “panes” daily, and they had the liberty of choosing “utrum per diem velint octo fercula, vel hiis ad certum tempus de coquinario integram et equivalentem alicujus piscis recipere porcionem” (pp. 73, 79, 98). This sounds like four corrodies (see J. W. Clark, The Customs of Augustinian Canons, Cambridge, 1897, p. 216; and Obedientiary Rolls of St. Swithin's, Winchester, ed. G. W. Kitchin, Hampshire Record Society, 1892, pp. 159 ff.). Some of the corrodies quoted by Kitchin are received “in aula domini Prioris” (p. 161).

17. Dugdale, v, p. 619.

18. He says this in connection with what he has to say about gluttony, staring, etc., etc. (see Morton, pp. 50, 68, etc.).—The French gives the text of p. 416 in a slightly superior reading (f. 67v).

19. It is perhaps noteworthy, in view of the injunction (p. 416) given by the author of the Riwle against keeping cattle, that at the Reformation Kilburn Priory seems to have possessed only one beast—a horse “of the color of black” (which, at least in the later days, when the favor of Westminster did not appear to be so personal as in the earlier, would doubtless be necessary for fetching the supplies from the abbey). Nevertheless the nuns cultivated forty acres of land in Willesdon. The prejudice of the author against the keeping of cattle by anchoresses, lest they should thus be drawn into quarrels because of complaints as to injury done by their beasts, seems to be typical of his time. The monks of the eremetical order of Grandmont, founded just before this time, were allowed to keep no living creatures except bees, “que vicinos non ledunt” (W. Map, De Nugis Curialium, Anecdota Oxoniensia, xiv, 1914, p. 26). These monks were all inclusi.

20. One of the many parallels which are to be seen between Kilburn and the nucleus of Sempringham appears at this point. Gilbert took the title of “master”—one of the points in which he imitated Robert of Arbrissel, the famous hermit who in 1101 established the famous double house at Fontevraud (see Migne, clxii; Fontevrault et ses Monuments, by Édouard, Paris, 1873). It was also the title of the head of the Templars, whose Rule—framed under the influence of St. Bernard of Clairvaux—was approved in 1128 (see the edition by H. de Curzon, Société de l'Histoire de France, 1886). The date of the beginning at Sempringham can be put 1131-5. The contemporary life of St. Gilbert mentions “the time of Henry I,” and we know that Gilbert returned from Lincoln about 1130 (Rose Graham, St. Gilbert of Sempringham and the Gilbertines, London, 1901, p. 10).—It should be noted that the V. C. H. (Bedfordshire, i, p. 539) speaks of a “warden or master as usual in small nunneries.”—The question of Godwyn's authorship of the Riwle is now being investigated, with promising results.

21. Le Coulteux, Annales Ordinis Cartusiensis, Montreuil, 1888, i, p. 302.

22. Migne, cliii, cc. 685, 755 f. It may be noted that St. John the Baptist was the special patron of hermits and anchorites all through this period, as may be seen by reference to Miss Clay's Hermits and Anchorites of England. The famous hermit Godric of Finchale, already mentioned as contemporary with the first sisters of Kilburn, saw St. John in person take him under his protection, and Godric's chapel—to which people from all over England resorted—was dedicated to the hermit saint, though his inner oratory, where no one might enter except himself, was dedicated to the Virgin (see Surtees Soc., op. cit., passim).

23. Gervase of Canterbury, in his catalogue of religious houses (written after 1199), thus describes the order and dedication: “Prioratus Keleburne, Sanctae Mariae, Moniales Nigrae” (Opera, Rolls Series, 1880, ii, p. 426). This tells us very little definitely. “Black nuns” was the phrase often applied to Benedictine women, but the women of Kilburn were often in the records called “moniales” and the habit of Augustinian canons was black. A similar careless statement which seems to arise from the connection of Kilburn with Westminster is the reference to the “abbes of Kylbourne” in a fifteenth-century parish record (E. E. T. S., No. 125, p. 152).—The full document cited by Park as first giving the title “Augustinian” to the house is printed by Thorpe, Registrum Roffense, London, 1769, pp. 264 ff. The bishop of Rochester here appropriates to the nuns a church because of their excessive poverty, which he has investigated and found due to the duties of hospitality to which their position on the highway exposes them, without fault on their part. He speaks in the beginning of “religiose mulieres, priorissa et conventus monasterii monialium de Kilbourne, ordinis sancti Augustini, London. dioc.” Elsewhere he always uses the indefinite phrase “religiose mulieres.” At this time Kilburn seems to have received new vitality.

24. Fasciculus J. W. Clark Dicatus, Cambridge, 1909, pp. 186-217. See also K. H. Schaefer, Kanonissenstifter im deutschen Mittelalter, Stuttgart, 1907, which, though it deals specially with the very peculiar development of canonesses in Germany, gives some details of them elsewhere. Dr. Frere, in reply to a letter from me on the subject of Kilburn, writes as follows of the use of the term “canoness” in its connection: “The word is merely applied to people who could not otherwise be described. … There is nothing available on the subject of English canonesses: they were very few and obscure. … In any case, whatever is the influence behind, the women are clearly recluses, not forming a regular Order, not professed virgins, for they have no ring, p. 420 … the very indefiniteness of all this seems to point to the 12th century rather than the 13th.”

25. See J. Lanteri, Eremi Sacrae Augustinianae, Rome, 1874. The term that is usually applied to the nuns of Kilburn in the charters is “ancilla Christi,” a phrase of very long descent and sometimes indefinite meaning (see Dictionnaire de l'archéologie chrétienne et de liturgie, ed. Cabrol et LeClercq, Paris, 1907). It is used for the nuns of Fontevraud, Marcigny, and Sempringham, and appears in the office for enclosing an anchorite in the York Manual (Clay, p. 193). It is apparently rendered as “mayde cristes” by Thomas de Hales in his Love Rune (E. E. T. S., No. 49, p. 93).—As a matter of fact, the Pope in 1148 decreed that “sanctimoniales & mulieres, quae canonicae nominantur, & irregulariter vivunt, juxta beatorum Benedicti & Augustini rationem, vitam suam in melius corrigant & emendant” (Mansi, Sacrorum Conciliorum Nova Collectio, Venice, 1776, xxi, c. 714).

26. The hermits about Cluny owed “obedientiam” to the monastery (Peter the Venerable, loc. cit.), and from such arrangements grew up the use of the title “obedientia” to designate cells. The character of some of such establishments may be understood by Peter's complaint that he cannot send to his friend Guigo of Carthusia a certain book because it has been eaten by a bear in one of their “obedientiae” (c. 103). The isolation of these retreats evidently brought on moral dangers, for Abelard refers to the irregular life lived in “obedientiae” (Opera, Migne, clxxviii, c. 265).—Roger, the hermit under whom Christina of Markyate began her religious life, was a monk of St. Alban's giving his “obedientiam” to the monastery (De Gestis, p. 97), and it would be expected that Godwyn, the master of Kilburn, would be a monk of Westminster. He is, however, nowhere so designated in the records, and we shall see that in 1231 the master of Kilburn is a secular priest. Almost anyone might occupy a hermitage as a beadsman of the abbey. Grimlaic makes provision for outsiders becoming anchorites under the protection of a monastery (c. 596).

27. See Morton, p. 22, French text, f. 5, J. W. Clark, op. cit., pp. xcvii, cii.

28. This is to me the most plausible interpretation of this difficult passage. Mr. Macaulay has apparently understood it somewhat more literally, though he evidently finds it uncertain (p. 463). I do not offer my suggestion as a final one, though the evidence drawn from Kilburn would seem to make it very satisfactory. In any case the rigid identification with early Dominican convents, made by Father MacNabb (Modern Language Review, xi, p. 4), seems impossible. Later research may clear up this reference. For Shrewsbury recluses living in community, see supra, p. 481, n.

29. See Flete, p. 101; Widmore, pp. 37, 70.

30. Park makes a flagrant error when he refers (p. 171) to the prioress as the “senior person” who is to be put over the nuns, according to the foundation charter. This “seniorem idoneum” whom the convent of maidens are to choose after the death of Godwyn, is certainly represented by the “secular priest” of 1231.

31. In 1377 the bishop of Rochester states that the nuns wish to increase their numbers (see infra, p. 490, n).

32. It is certain that at times Sopwell held as many as nineteen nuns (V. C. H., p. 422).

33. Carthusians did not originally leave their cells for mass every day (see Catholic Encyclopedia). Some statutes made about 1120 for a Carthusian house in Calabria recognize two classes there, “coenobitae,” and “anchoretae.” The latter at times leave their cells, though they are to preserve silence when so doing (Le Couteulx, i, p. 237). They even at times eat in the refectory, and the sisters of the anchorage may have done the same, though we have no information on the subject. The author, however, draws the line at their eating with guests “outside” (p. 412), though he says that there are anchoresses of his day who do so. They hear mass every day (p. 262).

34. Though Mr. Macaulay generally confines himself to description and avoids generalizations and hypotheses, it is of interest for the theory here put forward of the connection of the additions of B with the original anchorage that he notes that some of the additions are “quite in the spirit of other passages in the Ancren Riwle.” He gives no opinion as to the origin of B, though he declares the new material to be interpolated.

35. Le Couteulx, i, p. 348.

36. Two incidents should be noted in this connection, though our evidence is not complete. We are told that the abbot who was deposed in 1213 was charged with incontinency, among other crimes. Widmore is inclined to doubt the charge (p. 34). Again, Sir Walter Besant relates (Westminster, London, 1895, p. 116) that the abbot and convent once all took refuge at Kilburn, because a prophet had prophesied a flood which would drown the abbey. No date or reference is given.

37. Anglia, xxx, p. 116. See, for a refutation of a theory here propounded by Heuser, Modern Language Review, iv, pp. 433 f. Heuser seems to take it for granted that B was written for Wigmore Priory in Hereford—which procured it, as Mr. Macaulay shows, only about 1300 (p. 145).

38. Polychronicon, Rolls Series, viii, 1865, p. 190; Peter of Blois, Opera, ed. J. A. Giles, 1847, i, p. 350 (a letter to Geoffrey, Archbishop of York, apparently on his accession). With William de Newburgh, (Historia Anglicana, ed. H. C. Hamilton, London, 1856, i, p. 121) compare W. Map (op. cit., p. 56). It is noted that the first mention of heresy in London occurs in 1210 (V. C. H., London, i, p. 185). For the Continental heretics and the controversial works they called forth, see Vacandard, Vie de St. Bernard, Paris, 1897, ii, pp. 209 ff., E. S. Davison, Some Precursors of St. Francis of Assisi, Columbia University dissertation, 1907, p. 29.

39. E. Bishop, On the Origin of the Feast of the Conception of the Blessed Virgin Mary, London, 1904; The Bosworth Psalter, London, 1908, pp. 43 ff.

40. C. Q. R., p. 349.

41. Our Lady's Dowry, London, 1875, 2nd ed., pp. 68, 251. He quotes from a correspondence of 1174-80 between Peter Cellensis and a monk of St. Alban's, in which a very general acceptance of the feast in England is implied. The principal evidence of its celebration in France comes from Fécamp, the confederate of Westminster (Vacandard, Revue des Questions historiques, lxi, pp. 166 ff.).

42. Printed E. E. T. S., No. xiii. I regret that I have no rotograph for the French at this point.

43. See infra, pp. 507, 512, 536. The heroines of these legends address the Saviour in just the type of mystical devotion described in the Riwle in the abstract.

44. C. Q. R., p. 339. J. H. Round, however, notes that earlier it is unsafe to take the name of Alfred as certainly English (Feudal England, London, 1895, p. 327). The same may be true of those involved here.—N tells us (p. 192) that the anchoresses were sisters by blood.

45. See William of Malmesbury, op. cit., passim. Some circumstances seem to suggest that the nuns of Kilburn are, as it were, a special memorial to Edward. Gervase, in the charter already quoted, gave the corrody of Ailmar “for the good of his soul and for the soul of King Edward, as well as for all his successors”; and the same purpose is given by Flete in his account of the foundation. At the time of the canonization of Edward, the nuns of Kilburn are given a confirmation of their benefits (Flete, p. 94). It should also be noted that when Edward III in 1353 grants the nuns of Kilburn exemption from all taxes, he calls their house “of the foundation of the king's progenitors” (Patent Rolls, Edward III, pp. 250, 539). His favor was probably procured by the effort of Simon de Langham, for this “most eminent of all our abbots” had a special affection for Kilburn (see Widmore, pp. 99, 187), and Edward III established a chantry at Kilburn for his soul (Thorpe, loc. cit.).

46. Though they are not formally described as lay-sisters they are evidently given over to their superiors much as if they were such. In the first settlement at Sempringham the seven maidens were at first waited on by village women, until William, the first abbot of Rievaux (founded 1131), persuaded Gilbert to add lay-sisters who should be under a rule. When the Gilbertine rule was drawn up, the part for the lay-brothers merely copied the Cistercians, and in general the special development of the lay-brothers seems one of the contributions of the Cistercians at this time, though these were, of course, not their invention (see E. Hoffmann, Das Konverseninstitut des Cisterzienserordens, Freiburg, 1905). It was apparently a part of the unusually humane and successful treatment of the lay-brothers by the Cistercians that the sermons of St. Bernard were translated into the vernacular for their use—so early that there has been some discussion as to whether the French text might not be the original (see L. Bourgain, La Chaire française au XIIe siècle, Paris, 1879, pp. 187 ff.; Vacandard, i, p. 459). M. Demimuid in his Pierre le Vénérable (Paris, 1876, p. 269) notes the influence of the lay-brothers on the rise of vernacular literature.—It should be noted that at this time when lay members made an important element in so many orders, it was sometimes difficult to keep this part of the community contented. The order of Grandmont, already mentioned as inclusi, was nearly wrecked in the middle of the twelfth century by a resurrection of the lay-brothers, and the Gilbertines had serious trouble with a similar insurrection (see Walter Map, loc. cit.). The latter did not generally work with their laymen, as did the Cistercians, but it has been thought to be on account of offense to the lay-sisters that the nuns were forbidden to speak Latin. Peter the Venerable took care about the instruction of the lay-brothers of Cluny (Bourgain, p. 185).

47. Heuser declares that no manuscript of the work can be localized at Tarente, where its composition was traditionally placed, and R. Jordan (Die mittelenglischen Mundarten, Germanisch-Romanische Monatsschrift, ii, p. 128) agrees, as does also F. Weick (Aussterben des Präfixes ge- im Englischen, Darmstadt, 1911, p. 93), who is led by the partial loss of the sense for the ge-, manifested in the Riwle, to decide that it must have had a “Mercische Urform,” which had been the conclusion of Heuser. Jordan comes near the truth (as shown by the present hypothesis) when he conjectures that the treatise was composed in Oxfordshire.—It is a fortunate circumstance that we have in the Poema Morale (put for its place-names very securely in Hampshire, and by its manuscripts no less surely before 1170) an excellent touchstone for the locality of the English text of the Riwle, if written at Tarente. It is probably the decisive factor actuating the opinions just quoted against a Dorset origin for the treatise. It ought also to be remembered when the traditional dating of the work after 1200 is considered.

48. A thorough study of the language of the “Katherine group” has been made by H. Stodte (Ueber die Sprache und Heimat des Katherine-Gruppe, Göttingen, 1896). It is interesting that he notes one point of phonology found in the Group as also peculiar to London (p. 27), and a syntactical peculiarity is noted by the editor of the legends as like the usage of the “modern dialects of Middlesex and Wilshire” (E. Einenkel, E. E. T. S., No. 80, p. 135).

49. Morsbach calls it “einen im ganzen südlichen Dialekt, der vom Mittellande beeinflusst ist” (Ueber den Ursprung der neuenglischen Schriftsprache, Heilbronn, 1888, p. 162), and Sir James Murray “more southern than anything else, with a slight midland admixture” (Encyclo. Brit., art. English Language, p. 594). E. Dölle has extended the study of the London dialect behind the period treated by Morsbach (Morsbach's Studien, xxxii, Zur Sprache Londons vor Chaucer), and E. Neufeldt makes a study equally interesting for our purposes in his Zur Sprache des Urkundenbuches von Westminster, Rostock dissertation, 1907 (this is the manuscript Faustina A III, from which Dugdale prints the Kilburn charters). All of Neufeldt's material is much earlier than our period, and Dölle's either much later or much earlier, except Group VII—charters of the time of Archbishop Theobald, 1142-60.—A peculiarity of the Westminster charters noted by Neufeldt (pp. 44, 92) is the almost consistent use of a-forms for the verb habban. They seem to predominate in the Riwle.

50. Considering the early date of the material used, it is not strange that the Midland influence seen in the Proclamation and in the Riwle (which, though composed earlier, only exists in manuscripts written after 1230) is not yet visible in the documents examined by Dölle (p. 88). It should also be noted that the Riwle, in the use of u for A.-S. y, shows a more Western speech than the London dialect, as shown in the legal documents (Dölle, p. 26). However, cases are found in the Westminster charters (Neufeldt, p. 52), and in the Proclamation (Morsbach, p. 161-2), and the truth is that the cosmopolitan position of the city (which was ultimately strong enough to shift its dialect entirely) must have brought in an unusual number of speech mixtures. It is interesting to note in this connection that Heuser declares that the dialect of the Riwle is such a mixture that the title “South-Mercian” would suit it better than “Southern,” since so many exceptions have to be made if the latter title is used (p. 113). Morsbach, in listing the peculiarities of all the Middle English dialects in his Grammatik (Halle, 1896, i, p. 9), says: “Die Katherine-gruppe bildet eine gruppe für sich.” It is possible that the mixture of dialect which does not appear in the earliest London legal documents, would be visible in literary performances of the same time. It is also barely possible that anything written at Kilburn (which, though so near the city, was at this time a very rustic situation) might show more of the local Middlesex peculiarities of speech—though our ignorance as to the origins of the scribes, author, and public of the treatise (even when we are sure of the place of its composition) must make any such hypothesis uncertain, especially since our only copies date from so long after the original copy.—It should be noted that Father MacNabb, in the article already mentioned, quotes from a letter of Mr. Macaulay (p. 5 n.), in which the original dialect is spoken of as “Southwestern, with a tinge of Midland.” It would appear from Mr. Macaulay's statement made in his article and already quoted, that “Southwestern” here must be a scribal error for “Southern.”

51. See O. P. Behm, The Language of the Later Part of the Peterborough Chronicle, Göteborg, 1884, pp. v, 19, 20, 24, 57, 66; H. Meyer, Zur Sprache der jüngeren Teile der Chronik von Peterborough, Jena, 1889.

52. For example, most of the contents of Bodl. ms. 343 (History of the Holy Rood Tree, E. E. T. S., No. ciii, ed. Napier, p. ix).

53. The phonological changes in the Riwle are briefly summarized by E. Buck, Transactions of the Philological Society, London, 1865, “The Grammatical Forms of Southern English (A. D., 1220-30) occuring in the Ancren Riwle,” p. 166. In addition to the philological works already cited, the following may be mentioned as giving valuable comparative data and full bibliographical references: Anecdota Oxoniensia, Old English Glosses, Oxford, 1900, and Some Points of English Orthography in the Twelfth Century, Academy, 1890, pp. 133 ff., both by Napier (the latter lists manuscripts transcribed in the twelfth century); W. Schlemilch, Beiträge zur Sprache und Orthographie Spätaltengl. Sprachdenkmäler der Übergangszeit (1000-1150), Morsbach's Studien, xxxiv (a very useful study for the present purposes, summarizing the results of most of the other philological investigations here noted). The study by R. A. Williams (Anglia, xxv, pp. 393 ff.) of the speech of the Codex Wintonensis, drawn up in the bishopric of Henry of Blois (1130-50), later to be mentioned as a person of interest for the Riwle (infra, pp. 526 f.), is also useful for its summaries of other contemporary dialects. It should be noted that Group VII of Dölle's study (written in the time of Archbishop Theobald, who is also of interest for the Riwle—see infra, p. 535 n.) shows (pp. 5, 54, etc.), the French influence on the orthography very strongly (ch for c, etc.), which is such a characteristic of the Riwle, and is cited occasionally in the dissertations. Group VII only once (p. 43) substitutes i- for the prefix ȝe- (a sign of late dialect found consistently in the Riwle). Dölle notes, however, that legal documents are naturally conservative (p. 45), and the i- is found in some twelfth-century mss. (Academy, pp. 133 f.), and it even appears in the eleventh-century glosses (Anecdota, p. xxviii, in the work of a Kentish scribe). The Peterborough Chronicle derives much of its archaic appearance from its use of the prefix ȝe- —as a matter of fact it does not make such difficult translation as the Riwle—but it alternates this with entire loss of the prefix, according to the Northern custom (Meyer, p. 86), and altogether the form of the prefix seems a matter determined more by the locality of composition than by the date. Liebermann prints a short original charter written before 1128 in the North, which shows traces of Anglo-French orthography and of Middle English forms and the use of the “W.-S. Schriftsprache” (Archiv, cxi, p. 284).

54. For references see Schlemilch, p. 69.

55. Opera, Rolls Series, 1868, vi, pp. 177-8. I have pointed out in an article in the Romanic Review, ix, pp. 154 ff., which supplements the present one at many points, that our conception of the position of English at this time has been influenced by the fact that Giraldus, who speaks thus impersonally of English, was a Welshman, and that the fact that he and two others of the most outstanding writers of the time were Welsh probably had its influence in hampering the development of English as a medium for literature.—A use of isaid (showing the ge- worn down to i-) occurs in an English proverb quoted by Giraldus at p. 187. Part of one of his proverbs occurs in the Riwle (p. 268), and makes difficulty for the theory of a French original.

56. See my article already mentioned, pp. 186 ff. Attempts have been made to attribute all these pieces to the same author—according to Einenkel, loc. cit., and in Anglia, v, pp. 265 ff., a woman (perhaps one of the three anchoresses).

57. See W. Vollhardt, Einfluss der lateinischen geistlichen Litteratur …, Leipzig, 1888, p. 19. He refutes Einenkel's theory of the origin of these pieces.

58. See Morsbach, Grammatik, i, pp. 9, et passim.

59. R. Mattig (Die französischen Elemente im Alt- und Mittelenglischen, Marburg, 1910, p. 79) shows that half of the romance loan-words found in Middle English from 800 to 1258 are contributed by the Ancren Riwle.

60. See the valuable article by Mr. Henry Bradley in the Cambridge History of English Literature, ii, p. 436 ff.

61. Romanic Review, loc. cit., p. 173.

62. Op. cit., p. 45. Reyner (Apostolatus Benedictinorum in Anglia, Douai, 1626, pp. 159-60) says that the order of Arrouaise came to England in 1112, that of Tyron in 1126, that of Savigny in 1124, the Premonstratensians (who were in Scotland in 1125) to England in 1146. The first Cistercian house was built in 1128, but the real strength of the Cistercian order began with the settlement of Rievaux directly from Clairvaux in 1131 (see Miss Cook's most valuable account of the Cistercian Settlement in the Eng. Hist. Rev., loc. cit.). The founders of Fontevraud, Tyron, Savigny, and Arrouaise had all been hermits.

63. Migne, clxxviii, c. 213. For the dates of Heloise's letters see the Catholic Encyclopedia.

64. A useful contemporary account of the Cistercians for our purposes is that of Ordericus Vitalis, Historiae Ecclesiasticae, ed. A. Le Prevost, Société de l'histoire de France, Paris, 1845, iii, p. 434 ff. Ordericus is a “black monk,” a Benedictine under the influence of Cluny, who gives, therefore, a somewhat sarcastic account of the “white monks.”

65. The question of the color of the habit was a burning one at the time. The monks of Bec also assumed white, though the date is not known (Porée, i, p. 505), and the Premonstratensians wore it (explaining their choice by a miracle, Dugdale, vii, p. 859). The Gilbertines showed their complex origin by combining the two colors (Graham, p. 70). The author of the Riwle did the same (see Mac-Nabb, p. 4).

66. Ep. I.—It was put first in Bernard's collected letters because of the miracle connected with its composition, and a chapel existed on the site of the event, until after the Reformation. See Vacandard, i, Chap. vi, Demimuid, op. cit., etc. Barnard gave an Apologia for his attack on Cluny, which is really a second onslaught. It is dated by various scholars at dates ranging from 1119 to 1127 (the latter is that given by Miss Cooke, who gives an excellent summary of the causes of the Cistercian separation from the older Benedictines).

67. Migne, clxxxii, c. 906.

68. See Newman's Lives of the English Saints, London, 1845, No. 1 (Stephen Harding), p. 54, for a discussion of the “frock” and the “cowl,” and Migne's note on the letter to Robert.

69. See Gilbert Crispin, p. 27. Almost half of the houses of Savigny were in England (Cooke, p. 669). For their habit, see Dugdale, v, p. 246.—See Die Winteney Version der Regula S. Benedicti, ed. Schröer, Halle, 1888, p. 111.

70. He remarks on the shortness and scantiness of the Carthusian garments (c. 944). The number of the Cistercian garments was also restricted; see William of Malmesbury (p. 382), and Peter's second letter (c. 328). So were their kinds of bed coverings (ibid., p. 113).

71. See E. Bishop, “The Origin of the Cope as a Church Vestment,” Dublin Review, Jan. 1897, p. 24. A “cope” was generally the particular mark of a Canon—see Clark, p. lxxix, and Ordericus Vitalis, iii, p. 369; “Clerici et episcopi nigris cappis induti erant. Monachi quoque et abbates nigris nihilominus cucullis amicti erant.” Like all the costumes in question, it was also, however, worn by all classes, and at an earlier date, at any rate, was the term regularly used in some other countries where the French used “cowl.”—Matthew Paris in 1258 notes the arrival of eight ecclesiastics in England “in copes,”—“videlicet, quinque clausis, et quinque manicatis” (v. Ducange, Cappa). Rock (Church of Our Fathers, ed. Frere and Hart, London, 1905, ii, p. 41) quotes from Chaucer as to knowing a canon because “his cloke was sewed to his hood.”—It is very possible that we have in the Riwle itself proof that the author used “cowl” and “cope” interchangeably, as we have seen was sometimes done (most of these terms of costume seem to be sometimes interchanged, as examination of Ducange will show). … “Keppen” … has been taken to refer to head-coverings by Mr. Macaulay (p. 68), though Morton translates “capes.” I believe the latter may be preferable. We really have the same word as in “copes,” and an examination of the French text seems to show that we here may have it used in the same sense: “Si vous poez estre sanz wympel, seez od chaudes kuueles qe len appele kappes & par desus cel les noires veilz” (f. 68v.). Now “kuueles” is the word which we have used in the English text at p. 10, already quoted (supra, pp. 521 f.), where it is translated “cowl.” Since the “cowl” had a hood, a black veil could be worn over that, in lieu of a wimple.—Robert of Arbrissel also decrees for his nuns “ut guimpae albae earum nunquam apparent; velis eas operientibus” (Migne, clxii, c. 1079). The following from the Chaucerian Romance of the Rose has been quoted in this connection:

    Werynge a fayle in-stede of wymple,
    As nonnys don in her abbey

    (l. 3865).

    —The French text corresponding to the English at p. 11 is hardly legible: nothing can be recovered as to the “wide hood,” or the colors of the “cowl.” However, the latter word is partly visible, in the same form here found in the English and at f. 68v in the French (… ouele, f. 3v). The French abridges the reference to costume at p. 10 and omits the “kirtle or the cowl,” as follows: “… & le quel blank ou noir sicome nounsauanz vous demandent, qe quident qe ordre siete (?) en la cote.”—I regret that I have not the French text at p. 56.—The French also uses “cotes” for the lay-sisters' garb at p. 424: “Lour cotes soient par desus closes pardeuant la poitrine sanz fermail” (f. 69). N here uses “hesmel,” but B and C “cop” (Macaulay, p. 331). It would appear that the “ilokene cope” at p. 56 might not differ from the “cote close” of the lay-sister at p. 424 (as given in the French): thus, as directed (ibid.), her costume showed her dedication. “Cote” in the French often translates “kirtle” (as p. 362). Ducange gives instances showing “cotta” in monastic use, and one from 1298: “In decenti habitu, scilicet, in cappa clausa, vel cotta.”—The interpretation of such terms in the present treatise is specially difficult because we have two versions, dating from two widely separated periods.

72. G. G. Coulton, Mediaeval Studies, 1st Series, 2nd ed., London, 1915, p. 40.

73. Henry of Blois (nephew of the king and the dominating figure in the kingdom after the accession of his brother Stephen) was a monk of Cluny and one of its greatest benefactors. He became abbot of Glastonbury in 1126, Gilbert Foliot, Prior of Cluny, became abbot of Gloucester in 1139, the 12th abbot of Ramsey in 1176 became the 14th of Cluny, etc.

74. There were three degrees of dependence on Cluny (see Pignot, ii, p. 313 f.). Hugh, the first Abbot of Reading, was called there from the Cluniac house of Lewes. He collaborated with Osbert of Clare in establishing the Feast of the Immaculate Conception (on which Peter was skeptical, see Demimuid, p. 207), and, as Archbishop of Rouen, was at the death bed of Henry I. He was a relative of Peter's intimate friend, Cardinal Matthew (the reforming prior of Cluny), who as Papal legate in France came into constant conflict with the Cistercians (see Revue Bénédictine, xviii, 1901, p. 129).

75. The Anglo-Saxon Chronicle says that he was received everywhere with great honor (ed. Thorpe, Rolls Series, 1861, i, p. 380). See also Reyner, App., pp. 140 ff. Peter mentions in his letters two visits to England, the last of which was certainly after 1141, when he went to Spain (cc. 661, 671). He writes an undated letter to Henry of Blois in which he says he spent much time in England the previous year, almost constantly in Henry's company (c. 204). The position of this letter in the series would seem to refer it to the earlier visit. The Histoire littéraire says that Peter's second visit took place c. 1145 (xii, p. 355). It should be noted that the great basilica of Cluny, so important for the history of architecture, was dedicated in 1131, and for this Henry I had been the principal benefactor. Peter writes to the Empress Matilda that her father had been a better friend to Cluny than any king for three hundred years (see Duckett, Record Evidences among Archives of the Ancient Abbey of Cluny, privately printed, 1886, p. 42).

76. Migne, cl., c. 443.

77. See p. ix, n.

78. He once writes to Henry of Blois (c. 319) a begging letter for Marcigny, where the mothers of both had been nuns. Again he complains to Henry that the English visit Marcigny, and pass by Cluny (c. 231). He writes to one Robert who is apparently at Reading, and about to take the vows at Cluny, bringing with him a sister to enter Marcigny (c. 262). Marcigny had a dependent nunnery in the diocese of Salisbury (Pignot, ii, p. 41), but not even its name is known.—For Peter's love for Marcigny see cc. 208 f., 350, 457, 889.

79. The controversy was alive after the death both of Bernard and of Peter, as is shown by the Dialogus inter Cluniacensem Monachum et Cisterciensem, written from the Cistercian point of view between 1154 and 1174 (see Martène et Durand, Thesaurus Novus Anecdotorum, Paris, 1717, v, cc. 1569 ff.). This work is an elaborate discussion of each point in question between the two orders, and probably other details of the Riwle besides those mentioned are framed with reference to these arguments. For example, the Cistercians made a great point of saying matins at day-break (c. 1604), and then not going again to bed. Bernard taunts Robert with the “sweet morning sleep” at Cluny. The anchoresses may say matins “by night in winter” (p. 20). The Dialogus discusses the question of monasteries assuming control over religious women (on which the orders differ, cc. 1633 f.).

80. See Romania, xl, p. 745.

81. It is impossible to read the letter of St. Aelred without recognizing that some relation must exist to the Riwle. It has always been supposed that the latter was the borrower, because of the late date which it was given, and because of one citation “as Saint Aelred wrote to his sister” (Morton, p. 368, French, f. 59v). But any citation may be added by the scribe. We know that T and the Latin copy give extra texts (Morton, p. 184 f., Macaulay, p. 75), but Mr. Macaulay does not collate citations. In any case what is quoted is only the general sense of a section of the letter (c. 1460), and the mention of Aelred's name is therefore all the more likely to be simply the vague remembrance of a scribe. Aelred says (c. 1451) that he has no experience of the recluse's life, and writes from the works of “doctors.”—Some of the similarities found in the two works may be simply echoes of contemporary discussions. For example, the injunction against “gathering” is surely such (v. supra, p. 482 n.); and that against keeping a school, found in both treatises, may be a reminiscence of episcopal decrees; for it is quoted from “the Council of Rouen” in the same manuscript in which occurs Abelard's Rule for the Paraclete (Migne, op. cit., c. 322). I regret that I cannot trace its origin. The same excerpt forbids the reception of treasures for safe-keeping (also found in the Riwle). Both regulations are directed to “black nuns,” and the editor believes that they represent a compilation made by Heloise. Several councils of Rouen are listed from this period, but their decisions do not seem to be on record in the books accessible to me.—The fact that the gossip spread at anchorages, mentioned by both treatises, appears in a proverb quoted by the Riwle (p. 88) seems to show that this subject is not original with either writer. Probably St. Aelred's letter was not an early work, for his sister was old (v. cc. 1454-1457).—It is probable that the injunction against gay needle-work, found in the Riwle, originated in conditions which the author had observed. We know at least that Christina of Markyate did fine needlework “to gain friends by,” as the anchoresses are told not to do (p. 420): for she helped the abbot of St. Alban's by sending some of her work to Pope Adrian IV (therefore after 1154).—B adds that needlework may be done for pay, if the master permit, and need require, and C, in adding the same by later correction, adds that this should be arranged secretly (Macaulay, p. 331).

82. Migne, cxcv. He refers to being Master of the Novices (c. 562), and he left this position to lead out the house of Revesby in 1142-3. The work was instigated by the Abbot of Louth Park, which was not settled till 1140 (Cooke, pp. 662, 660).

83. Opera, ed. J. A. Giles, London, 1848, iv, p. 182.

84. Opera, ii, app., p. xcii (in an answer to a detractor, who is a canon who has left his convent to live as an anchorite, p. lxxxix).

85. See his refusal sent to the Bishop of London, Ep. 123 (i, p. 371). He makes much here of the fact that the Carthusians do not celebrate mass frequently (p. 373).—It is possible that the suggestion made to the anchoresses (p. 286) to “vren lesse uorte reden more” (noted by Mr. Macaulay, p. 73, as unusual) shows the Carthusian influence elsewhere apparent in the Riwle. The Carthusians were famous for their love of books, and Guigo in his Customs says that almost all are scribes (c. 694).—“Aux yeux du Clunisien, le moine était avant tout l'homme de la prière, presque uniquement l'homme de la prière liturgique” (Rev. Bénédict., loc. cit., p. 285). The Cistercians reacted in favor of manual labor against the numerous masses of Cluny, but they of course made little of learning, and seem to have set prayer above reading. The Riwle may, by “vren,” mean saying extra hours (left to the choice of the anchoresses, p. 44). Peter was influenced by the Cistercians in reviving manual labor for the monks of Cluny, but for the anchorite Gilbert (cc. 97-98) he seems, by his praise of the copying of books, to show the Carthusian influence.—Another reference in the Riwle is certainly a contemporary echo. The anchoresses are told to keep silence at meals … (p. 68). Guigo (c. 738) says that Carthusians should be taught by the Cistercian example to keep silence at meals. Peter the Venerable in his Statutes (c. 1032) says that the monks of Cluny ought to keep silence at meals “because all others do so.” This is an instance which shows the truth—in spite of reaction against their influence in some regards—of the saying of William of Malmesbury that the Cistercians were in his time “a model for all monks, a mirror for the diligent, a spur to the indolent.”

86. The former opens the chapter already quoted from by the sentence, “Hypocritarum autem nomen et notam cautissime et fidelissime declinant Carthusienses” (p. 180).—His admiration for their lack of avarice leads him to dilate on those who “beg in order to get rich” (p. 185), and again shows the stock character of the injunctions on this point given to the anchoresses.—For Peter of Blois's admiration for the Carthusians, see op. cit., i, pp. 259 ff., 304 ff. (Eps. 86 and 97), iii, p. 47 (his dissatisfaction with most monks of his time appears ibid., pp. 25, 31, etc., 57).—Peter the Venerable's predilection for the Carthusians above all other orders is many times expressed in his letters (v. cc. 371, 412, 429, 478), and he visited them generally every year (c. 28). He also visited the group of hermits in subjection to Cluny (cc. 360 ff.), of whom Gilbert (to whom he wrote the epistle on the Solitary Life) apparently made one, and it must be remembered that Marcigny, the house of women in which he took such an intense interest (where he also often stayed) included anchoresses. It is evident that the knowledge of the Carthusians was spread all over Europe long before the establishment of the first English house in 1174, and Guigo's Customs would doubtless also be known (by the instrumentality of Peter the Venerable, if no other way).

87. See the Tablet, 1907, Oct. 19, 26, Nov. 2.

88. See Father Bridgett's History of the Holy Eucharist in Great Britain, new edition by Father Thurston, London, 1908, p. 182; Henry of Huntingdon, Opera, Rolls Series, 1879, p. 271.

89. See Rock, iii, pp. 258 n. ff., and Bridgett (Our Lady's Dowry, p. 176 f.), who combats Rock on its late introduction. H. Leclercq (Histoire des Conciles, Paris, 1913, v, 2nd part, App. iv) gives a very complete article on the Ave Maria, treating its use by the Dominicans, etc. He notes one Aybert (d. 1140) who used it constantly. This man (in Hainault) was first a hermit, then a Benedictine monk, then a recluse visited from near and far (see Acta Sanctorum, April, i, pp. 627 ff.).—The use of the Ave Maria in the present case would certainly cause no difficulty, for Aves make part of the devotion of the “Five Psalms of the Virgin,” pp. 38 ff., first pointed out in the treatise by Father MacNabb (p. 3), and this devotion can be carried back to the middle of the twelfth century with the Aves as it is found in the Ancren Riwle (the form used by Jordan of Saxony, noted by Father MacNabb, does not contain the Aves; see Monumenta Ord. Fratr. Praedic. Hist., Louvain, 1896, i, p. 118). In one case we are told that the monk, on whose death (in 1163) a miracle followed connected with this devotion, was taught the custom of saying the psalms by Theobald, Archbishop of Canterbury, when he once stopped at St. Bertin's Abbey, St. Omer, on his way home from Rome.—Theobald went to Rome for his pallium in Jan. 1139, and this journey may be in question (see Ward, Catalogue of the Romances in the British Museum, London, 1893, ii, p. 633).—He said he had heard in Italy that the custom was practised in Jerusalem. In this case it may have been brought home by any traveller. It should be noted that the daughter-in-law of “good Queen Maud” (later Abbess of Fontevraud) was the daughter of the King of Jerusalem, and that Marcigny had a daughter-house there (Cucherat, Cluny au XI^e siècle, Autun, 1873, p. 92 n.). The church of the Holy City had its own liturgy, which was at some points very advanced in development (see J. Wickham Legg, Essays Liturgical and Historical, “Society for Promoting Christian Knowledge,” London, 1917, pp. 157 ff.). Jerusalem and crusading customs were doubtless of great interest at the time of the foundation of Kilburn.—A child is not taught the Ave in the time of the Riwle (see p. 210.—Cp. Rock, loc. cit.).

90. Several writers note how unsafe it is, in the case of devotional practises, to date the original use of a custom at the time it first appears in the documents. In this connection should be quoted the letter of Dr. Frere already mentioned: “The general character of the services seems to me to point rather to the 13th century than to the 12th, but this is not really cogent, because one knows very much less of the services of the 12th century than those either of the 11th or 13th.” I wish to thank Dr. Frere both for his kind letter, and for permission to use it here. He wishes me to make clear that he gives “only hasty impressions.” He makes no conclusion.

91. That in the Romanic Review, already mentioned, in which I group together the mystical works found in England in all three languages, and connect them with the historical evidence available which would seem to give hints as to their origin.

92. It should be noted that Kilburn by the foundation charter was to pray for the abbey of Fécamp, and it had been the Abbot of Fécamp of an earlier time who had circulated Anselm's meditations (see my earlier article, p. 184 n.). A female recluse of Fécamp signs the mortuary roll (1122) of Vital of Savigny (Rouleaux des Morts, ed. L. Delisle, Société de l'Histoire de la France, 1866, p. 296, 281 ff.). Fécamp was to the dukes of Normandy much what Westminster was to the kings of England. Edward the Confessor spent his exile here, and a close relation with England was kept up because this was the port of embarkation for England. Abbot John of Fécamp translated his interest in mysticism into practice, for he helped establish two monks as anchorites (L. Fallue, Histoire de Fécamp, Rouen, 1841, pp. 105, 129, et passim).

93. This, we must remember, was the home of the scribe of T, which, as we should also remember, contained others of the mystical pieces (see Mühe, Ueber den im MS. Cotton Titus D XVIII erhaltenen Text der Ancren Riwle, Göttingen, 1901). The Riwle in T is adapted for both sexes (Mühe, pp. 48, 50).

94. Some sentences are identical in both works (v. Vollhardt)—which presents difficulties for the theory of a French original for the Riwle.

95. E. E. T. S., No. 80, p. xvi.

96. It is placed between 1135 and 1147 (see edition in the Rolls Series, 1888, ii, introd., i, p. xviii).

97. Mr. Macaulay implies (p. 473) that the “anchorite sisters” of Ghent for whom he made this version could not belong to the house of Tarente, which was Cistercian. But the fact that anchorites lived under the protection of religious houses must be taken into account, and it may be that the term “apud Tarente” refers to residence near the Abbey in some such capacity.—It is interesting that Bec, in the abbacy of Anselm (who was their special friend), took under its protection three women who lived as recluses outside the cloister 1079-99. Two of these were great ladies retired from the world, of whom one was the mother of Abbot Crispin of Westminster (see le Chanoine Porée, Histoire de l'abbaye du Bec, Evreux, 1901, i, pp. 182-3). One may wonder whether the corrody of Gilbert may not have been given to the ladies of Kilburn in memory of the earlier trio, of whom his mother had been one. It will be recalled that Abbot Herebert has been thought to have come from Bec with Crispin.—There were regular Dominican recluses (see Clay, p. 78).

98. The part relating to the lay-brothers is lacking in the Westminster Customary as we have it, and the hours given in the related Customary of Canterbury (p. 281) are not those of the Riwle.—The author of the Riwle says (p. 6) that the unlearned must say their hours “in other wise.”

99. On the identity of these terms see the Oxford Dictionary.—To match Father MacNabb's reference to the special devotion to Mary Magdalen evident in this passage, see St. Anselm's veneration of this saint, noted in my former article (p. 188 n.). The connection which Father MacNabb seeks to make between the use of the term “the order of St. James” and the title “Jacobitae” given to the Dominicans, is surely unnecessary, since the author has given his own explanation of the phrase. If a cult of St. James is needed with which Kilburn can be connected, it can be found at Reading Abbey, where one of wide patronage developed on Henry I's presenting the house with a relic of St. James the Great in 1125, or thereabouts. See the history of the Abbey by J. B. Hurry, London, 1901, pp. 95, 130, 163. It was St. James the Great whose name is given to the Dominicans (v. Ducange), but the epistle which gives the title to the anchoresses belongs to St. James the Less.

100. See Matthew Paris, Historia Anglorum, Rolls Series, 1866, ii, pp. 397, 405, Dugdale, v. pp. 619 ff. The place came to be known—even in charter—as “Locus Reginae super Tarent.”

101. See W. Wallace, Life of St. Edmund of Canterbury, London, 1893, p. 352. The interest in the religious education of women in the thirteenth century was surely not so remarkable as that in the twelfth. This will be discussed in a later paper.

102. This passage may represent a real habit of thought of the house. The Virgin was considered to be the head of Marcigny, and her seat and portion were always provided (Pignot, ii, p. 35). It may be that the Carthusians sometimes took Christ for their head in the same fashion, for John of Salisbury mentions that they call no saint their founder, and adds: “Alii Basilium, alii Benedictum, hi Augustinum, at isti singularem magistrum habent, dominum Jesum Christum” (Opera, iv, p. 183).—The use of “distinctions” and of interpretations of Hebrew names, connected by Father MacNabb with Dominican influence (pp. 3, 6), means no more than a scholastic training in the author. This point will be discussed in a later paper.

103. See J. W. Clark, op. cit., p. lxvi.

104. Chronicle of Abingdon, Rolls Series, 1858, i, p. 49, ii, p. 464.

105. Gilbert Crispin, p. 31. It is perhaps a sign of the English character of the convent that the Westminster Customary uses the old English habit of holding a cup (p. 127).

106. See Clark, pp. 82, 98. The part which would treat this subject is missing from the Westminster Customary. The division into alternate choirs seems to have been a general monastic custom (Martène, De Antiquis Monachorum Ritibus, Lyons, 1690, pp. 26 ff.).

107. Peter the Venerable makes some alleviation in this regard in his reformed statutes of Cluny (c. 1043), and the Rule of the Temple criticizes the immoderate standing that had been practised (p. 26). These relaxations perhaps represent a reaction from the ideal of the eleventh century, which may be seen in Peter Damian. He views with the greatest horror the sitting during divine service which he had witnessed in France (op. cit., cc. 641 ff.).

108. They are mentioned in a chronicle written in the late twelfth century, printed in Dugdale, vi, pp. 128 ff.

109. Father MacNabb implies (p. 6) that the woman whose austerities are referred to may be Rich's mother, but it is a part of his theory to see Dominican influence in the Riwle (which puts it after 1221), the reference is made in the present, and Mabel Rich died about 1203 (Wallace, p. 69).