St Francis's writings

Francis's sanctity, the miracles attributed to his intercession and the diffusion of his cult in the western Church ensured that his words were treasured and carefully conserved, above all by members of his own order. Despite his protestations of simplicity and his self-effacing comments about the limitations of his formal education,Footnote ⁴ he left a substantial body of writings. The small quantity of extant autographs, consisting of two folios containing three short works, shows that Francis himself was capable of writing in Latin.Footnote ⁵ Apart from these autographs, Francis is known to have written at the chapter of 1221 to the minister general and friars of France an autograph letter no longer extant, and there are references to his requesting writing implements to be brought to him.Footnote ⁶ Recent scholarship has cast doubt on the earlier view that Francis's education was rudimentary.Footnote ⁷ There is also some evidence that he dictated his thoughts to his companions, notably to Leo.Footnote ⁸ In most of his letters Francis requested that his words should be preserved and disseminated, and he sometimes imparted his blessing to those who fulfilled his wishes in this respect.Footnote ⁹ Francis enjoined that the texts of certain of his writings, notably the Testament and the Regula non bullata, should be preserved by the ministers and other officials of the order without addition or subtraction.Footnote ¹⁰ Angelo Clareno no doubt records a view widespread among the fraticelli when he states that the saint's writings served to preserve his wishes against friars intent on thwarting his will.Footnote ¹¹ He further attributes all that Francis said in the Rule, the Testament and his letters to a divine source.Footnote ¹²

A corpus of Francis's major writings was established at an early stage. None the less, debate about the authenticity of certain minor texts continues.Footnote ¹³ The collection of undisputed texts was sufficiently well-established and authoritative to be of use to Thomas of Celano, Francis's first biographer, whose Vita prima of Francis dates from 1228–9. Indeed, recourse to Francis's writings was one effective means for the hagiographer to authenticate the portrait of his subject.Footnote ¹⁴

The Vita prima by Thomas of Celano cites the Testament on the subject of Francis's perception of lepers prior to his conversion and the prayer to be recited in all churches.Footnote ¹⁵ Quotations and borrowings from the Testament,Footnote ¹⁶ the Rule,Footnote ¹⁷ some of the letters Footnote ¹⁸ and the Canticle of Brother Sun were identified by the Quaracchi editors in their critical apparatus to the Vita prima.Footnote ¹⁹ Francis's writings are likely to have circulated among the ministers provincial and the friaries at a very early stage. The Rule (Regula vitae) was read in common by the friars each Friday,Footnote ²⁰ probably during the meeting of the domestic chapter (while the general constitutions were to be read each monthFootnote ²¹ and the provincial constitutions every two months).Footnote ²² The copying and citation of Francis's writings served to stimulate friars to imbibe the spiritual lessons of their founder. These writings were also regarded as a vital supplement to the Regula bullata, the version of the Rule which was approved by Pope Honorius iii on 29 December 1223, and as a guide to Francis's thoughts regarding a range of topics which were not explicitly addressed therein.Footnote ²³ The Vita prima cited the Admonitions as well as other works by Francis,Footnote ²⁴ while a Liber exemplorum alludes to Admonition 19.Footnote ²⁵

Copies of St Francis's writings, even those which concerned the domestic affairs of the Franciscans, were owned and read outside the order. Members of other religious orders and some secular clergy took a keen interest in the counsel proffered by the man whose body had been marked in a hitherto unprecedented manner by the stigmata, the five wounds borne by the Crucified Christ. One of the earliest quotations from the Admonitions comes from an anonymous Dominican preacher at Paris on 13 July 1231.Footnote ²⁶ It will be apparent that admiration for and interest in St Francis was not confined to the order which bore his name. Two of the earliest representations of the saint were painted in the Benedictine monasteries of Mont Saint-Michel and the Sacro Speco, Subiaco.Footnote ²⁷ The latter is a wall painting datable by an inscription in the chapel in which it appears to the second pontifical year of Gregory ix (21 March 1228–20 March 1229). The subject is described not as a saint but as Frater Franciscus and lacks a nimbus. These features suggest that the painting was completed prior to Francis's canonisation on 16 July 1228.Footnote ²⁸ In other words it may be dated between 21 March and 16 July 1228. Matthew Paris, the chronicler at the Benedictine monastery of St Albans, reproduced the complete text of Francis's Rule in his Chronica maiora, which he illustrated with images of the saint and one of his early companions, William, a barefooted English friar.Footnote ²⁹

The extensive research of Esser and Oliger reveals something of the circulation of Francis's writings outside the order. Copies came into the possession of Benedictines (Staatsbibliothek, Bamberg, ms B.V.35), Dominicans (Universitätsbibliothek, Basle, ms C.V.19), Cistercians (Bibliothèque nationale, Paris, ms lat. 2477), Augustinian canons (Stadtbibliothek, Trier, ms 476) and Carthusians (Universiteitsbibliotheek, Utrecht, ms 167).Footnote ³⁰ A late fourteenth-century catalogue of the library of the Benedictine abbey of Peterborough listed two copies of the Rule and one of the Testament.Footnote ³¹

The biographers who followed Thomas of Celano enjoyed access to fuller collections of Francis's writings, as the vitae from the 1230s attest. John of Perugia, the author of the Anonymus Perusinus, quoted the Testament Footnote ³² and alluded to the Rule and other texts. The Compilatio Assisiensis likewise cited the Testament.Footnote ³³ The last two sentences of the portrait in Thomas of Celano's Vita secunda of the ideal preacher quote from the Testament and the Letter to St Anthony of Padua. The conduct of Elias of Cortona, minister general, who was deposed by the general chapter of Rome in 1239, concentrated reflection on the Rule.Footnote ³⁴ The friars and their opponents cited the Rule during the polemical exchanges of the 1250s and later 1260s.Footnote ³⁵ One of the first extant commentaries on the Rule in 1241–2 cited the Testament.Footnote ³⁶ Along with the Testament, it continued to circulate in the first half of the fourteenth century. It was natural that the controversies involving the fraticelli should have focussed attention on Francis's writings, above all the Rule and the Testament.Footnote ³⁷

Borrowings from the Admonitions have been located in the Anonymus Perusinus,Footnote ³⁸ and in Thomas of Celano's Vita secunda.Footnote ³⁹ The Admonitions, like the Testament and the Rule itself, were quoted in debates regarding the fitting observance of the Rule, although the extracts or summaries tend to be too brief to check against the manuscript tradition. The Compilatio Assisiensis is an early exception to this pattern in its quotation of the last sentence of the sixth Admonition: ‘sancti fecerunt opera, et nos, recitando et predicando ea, volumus inde recipere honorem et gloriam’.Footnote ⁴⁰ It is noteworthy that the Compilatio Assisiensis, in including the words ‘et predicando’, is identical to the text of the manuscript published below and most other manuscripts.Footnote ⁴¹ The edition by Esser and Grau prefers the reading ‘recitando’ alone, but Paolazzi restores ‘et predicando’.Footnote ⁴² Bonaventure of Bagnoregio's Legenda maior reproduces a brief passage from the twentieth Admonition.Footnote ⁴³ Similarly, the third,Footnote ⁴⁴ eleventhFootnote ⁴⁵ and fourteenth Admonitions were cited in Angelo Clareno's Commentary on the Rule.Footnote ⁴⁶

The Admonitions

While some of Francis's letters were addressed to groups outside the order, such as the clergy, all the faithful, and lay rulers, many of his writings, including the Admonitions, were intended specifically for friars. However, there is no direct evidence concerning the circumstances in which the Admonitions were composed nor concerning their date of composition. The Anonymous Perusinus reports that Francis delivered admonitiones to the friars concerning the observance of the Rule during the order's chapters.Footnote ⁴⁷ The Legenda trium sociorum refers in similar terms to Francis's admonitiones in the context of a chapter held at S. Maria della Porziuncula.Footnote ⁴⁸ Although it cannot be claimed that these texts are referring specifically to the work by Francis known as the Admonitiones,Footnote ⁴⁹ it is reasonable to suppose that this work derives in whole or in part from addresses that St Francis gave at meetings of the general chapter of the order. This circumstance is perhaps reflected in the title of the Admonitions in an important manuscript of the works of St Francis: ‘verba sacre admonitionis … ad omnes fratres’.Footnote ⁵⁰ Jacques de Vitry, in a letter of October 1216, refers to the Franciscans' practice of congregating together once a year.Footnote ⁵¹ The Regula non bullata of 1221 ordered that a triennial general chapter was to be held at La Porziuncula and an annual chapter of the Italian friars was to be held in the same place.Footnote ⁵² Esser regards the Admonitions as a single composition;Footnote ⁵³ but if it is correct to associate the Admonitions with the general chapters, this suggests that they represent not so much a single work as a series of texts compiled over a period of years.

It has been claimed that the Admonitions are reportationes, in other words that members of Francis's audience wrote down his words when he addressed the chapters or on other occasions.Footnote ⁵⁴ However, there is more than one problem with this view. As we shall see in the context of a discussion of the headings in manuscripts of the Admonitions,Footnote ⁵⁵ the works of St Francis, including the Admonitions, show a remarkable fidelity to the views of Francis on the appropriate use of vocabulary, a fidelity which suggests that the texts of his works accurately reflect not only his thoughts but also the manner in which he wished them to be expressed. A further problem is that the manuscripts of the Admonitions do not contain sufficient differences in the text to suggest that they derive from different reportationes. We are left, then, with a single rapporteur, who was intimately acquainted with Francis's ideas and his preferred means of expression, and it seems reasonable to substitute for that hypothetical figure Francis himself. The argument appears to be reinforced by the conclusions of Giovanni Pozzi's study of the prose style of Francis. He stresses the homogeneity of Francis's writings, with reference in particular to the autograph works and the Admonitions. Footnote ⁵⁶

The teaching of the Admonitions is grounded in the New Testament, and the work contains numerous quotations from the New Testament.Footnote ⁵⁷ André Vauchez has plausibly suggested that an implicit theme of the work is the repudiation of contemporary heretical beliefs.Footnote ⁵⁸ Francis tends to begin each Admonition with a scriptural quotation, an allusion or image, which is followed by a practical application.Footnote ⁵⁹ The text also contains echoes of the Beatitudes articulated in the Sermon on the Mount (Matthew v. 1–16 and Luke vi. 20–6). The Admonitions often start with ‘Beatus ille qui’, matched by the lament for those who transgressed this norm of conduct, ‘Ve illi religioso’. This negative element implies a note of disillusionment with those friars who failed to act according to the founder's ideals. This may suggest composition in the years following Francis's decision to return from the east and to stand down from leadership of the order (1220). The majority of commentators do indeed place the Admonitions late in Francis's life, that is, in the 1220s. Robert J. Karris, for instance, dates them to the last three years of his life.Footnote ⁶⁰ On the other hand, Giovanni Miccoli, following Damien Vorreux, suggests a date as early as c. 1209–21.Footnote ⁶¹

Vices and virtues were central themes in the friars' programme of preaching, as ordained by chapter nine of the Regula bullata.Footnote ⁶² This preoccupation informs the Admonitions,Footnote ⁶³ which also reflect the involvement of members of the order in the implementation of the decrees of the Fourth Lateran Council of November 1215.

Although numerous manuscripts survive containing the writings of St Francis, only six of them are believed to date from the thirteenth century. Five out of the six include the Admonitions; in two manuscripts, the Admonitions are the only work by Francis, as is the case with the manuscript discussed in this article.Footnote ⁶⁴

Among the manuscripts of the writings of St Francis, Sacro Convento di S. Francesco (formerly Biblioteca Comunale), Assisi, ms 338, holds pride of place. Although the date of this manuscript is disputed, recent scholarship has tended to reinforce F. C. Burkitt's view that the section of the manuscript (quires 2–4) containing works by Francis dates from the mid-thirteenth century.Footnote ⁶⁵ Esser and Pellegrini in their detailed studies of the manuscript arrive at conclusions very similar to those of Burkitt regarding the date, Pellegrini arguing for a date of 1244–53.Footnote ⁶⁶ This manuscript and Biblioteca Medicea Laurenziana, Florence, cod. X, Plut. XIX dextr. contain two of the earliest known copies of the Admonitions. The latter, from the library of the convent of Santa Croce, Florence, is a miscellany of texts associated with the order and a variety of other items. This manuscript shows some independence in its use of headings in the Admonitions, and it divides the text in a slightly different manner. The Admonitions are also contained in Bibliotheca Apostolica Vaticana, ms Rossin. lat. 23 and in Bibliotheca Guarnacci, Volterra, ms 225, manuscripts which have been assigned to the later thirteenth century. The final manuscript to be mentioned is Grays Inn Library, London, ms 2, which has been dated to the late thirteenth or early fourteenth century. These early manuscripts do not present a uniform text of the Admonitions. Thus the Gray's Inn manuscript omits the first Admonition, as do some later manuscripts, perhaps reflecting a perception that it was a separate work on the inculcation of eucharistic piety. In the manuscript published below, on the other hand, it appears under the Admonitions' main heading, ‘Hec sunt verba sacre admonitionis …’, with the subheading ‘De corpore domini nostri Iesu Christi’.Footnote ⁶⁷ Doubts have been expressed in the past about Francis's authorship of the first Admonition, but the consensus is now that it is a genuine work of the saint.Footnote ⁶⁸

A newly recovered text of the Admonitions

A manuscript containing the Admonitions was sold at Sotheby's in London on 22 June 2004 (lot 54) and passed into private hands.Footnote ⁶⁹ Nothing is known about the provenance of the manuscript prior to the sale. The number ‘15561’ written in modern pencil at the end of the text (fo. 6v) may be a bookseller's stock number. The manuscript numbers only six folios, measuring 189 mm by 131 mm. The text is in two columns and is written in the same hand throughout. The scribe was not a masterly calligrapher: his hand is fairly rough, and the lines end unevenly. What survives probably once formed part of a larger manuscript, but the two sections of the manuscript in its present state appear to be complete. The first section (fos 1–4) contains extracts from the Liber scintillarum, a very popular collection of quotations from the Bible and the Fathers. It was compiled c. 700 by Defensor, a monk of Ligugé near Poitiers. This section closes with excerpts from chapter sixteen, De silentio,Footnote ⁷⁰ and chapter eighteen, De sapientia.Footnote ⁷¹

The Franciscans took some interest in Defensor's collection which was to be found in their libraries at San Fortunato, Todi;Footnote ⁷² the Sacro Convento, Assisi;Footnote ⁷³ Il Santo, Padua;Footnote ⁷⁴ and Santa Croce, Florence.Footnote ⁷⁵ Franciscans from the custody of Newcastle gathering information for the Registrum Anglie, a union catalogue of libraries, noted that the Liber scintillarum was available at the Augustinian Priory of Guisborough, North Yorkshire, and the Cistercian abbey at Newminster, Northumberland.Footnote ⁷⁶ The catalogue of the library of San Francesco Grande in Padua contains two copies of Defensor's collection.Footnote ⁷⁷

The second section of the manuscript (fos 4r–6v) contains the Admonitions. It is in the same hand as the first section and runs straight on from it without any vacant space. It is noteworthy that in the heading Francis of Assisi is not referred to as sanctus or beatus. It reads: ‘Hec sunt verba sacre admonitionis venerabilis patris nostri Francisci ad omnes fratres. In nomine Patris et Filii et Spiritus Sancti. Amen. De corpore Domini nostri Iesu Christi.’ The failure to describe Francis as a saint strongly suggests that the manuscript antedates his canonisation, which took place less than two years after his death (he died on 3 October 1226 and was canonised on 16 July 1228).Footnote ⁷⁸ News of this event must have spread rapidly within the order and elsewhere, and it is difficult to imagine that a scribe would fail to describe Francis as sanctus or beatus once he was aware that he had been canonised. In theory, a scribe writing after the canonisation, might have slavishly copied an earlier manuscript which did not describe Francis as a saint, but this seems unlikely.

The handwriting of the manuscript can be studied in the facsimiles which accompany this article. It is quite different from that of the celebrated collection of writings of St Francis and other Franciscan texts, Assisi ms 338.Footnote ⁷⁹ Among the distinctive features of the hand are the two elongated bars in the capital E, the form of the er abbreviation after s, which makes s(er)- at first sight look like sy, and the rather prominent horizontal stroke which precedes s. The final s is invariably long. In order to attempt to date the handwriting, it is advisable to compare it with the various published facsimiles of dated and datable manuscripts. We have not found illustrated in these catalogues any examples that strikingly resemble this hand. One of the closest that we have noticed is Bodleian Library, Oxford, ms Lat. liturg. f. 1, although the hand of the Oxford manuscript is rather more cursive in character. It is a Cistercian breviary of French origin of c. 1219.Footnote ⁸⁰ Two manuscripts in the Bibliothèque nationale, Paris, display similar script, lat. 469 and lat. 6901. The first contains the Postillae of Hugh of Saint Cher, is datable to the years 1230–48, and belonged to Clarinus, bishop of Carcassone 1226–48.Footnote ⁸¹ The other, which is closer to the privately owned manuscript, contains works by Rasis in the translation by Gerard of Cremona, is dated 5 June 1230, but has not been localised.Footnote ⁸² In another Cistercian breviary, now Bibliothèque municipale, Troyes, ms 1907, from Clairvaux and datable to after 1228, the name Francisci is similar in appearance to the same word in our manuscript.Footnote ⁸³ The script may also be compared to that of the Epistola ad clericos published with a facsimile by L. Oliger from a manuscript in the Biblioteca Vallecelliana, Rome (Cod.B.24).Footnote ⁸⁴ That manuscript is from the Benedictine abbey of the Sacro Speco, Subiaco, and Oliger argued that the text of the Epistola ad clericos was copied there from Francis's autograph or the original exemplar.Footnote ⁸⁵ This section of the manuscript is datable to the years 1219–38,Footnote ⁸⁶ and it is therefore close to the likely date of the manuscript of the Admonitions. One may note, for instance, the similarity of the words ‘sanctissimum corpus et sanguis domini nostri Iesu Christi’ in the manuscript of the Admonitions to the words ‘sanctissimum corpus et sanguinem domini nostri Iesu Christi’ in the Subiaco manuscript.Footnote ⁸⁷ Although it would be difficult to date our manuscript of the Admonitions on the basis of its handwriting alone, comparison of the latter with dated and datable manuscripts does not contradict the evidence of the title; in other words, the handwriting is compatible with a date prior to Francis's canonisation.

The first line of text is written below the top ruled line, with the exception of one page, where the words ‘et non plus’ are above the top line.Footnote ⁸⁸ N. R. Ker, in discussing English manuscripts, noted that in around 1230 scribal practice changed from writing above the top line to writing below it. He associated the new procedure down to the middle of the thirteenth century with the work of professional scribes. It must be emphasised that we are not here concerned with a sudden change and that the timing of the change may have varied in different parts of Europe.Footnote ⁸⁹ In the case of the present manuscript, all that can be said is that writing below the top line neither contradicts nor supports a date before Francis's canonisation.

A possible objection to the argument advanced here is that, to judge from the first edition of Esser, there are several manuscripts of the Admonitions which are undoubtedly of later date and which do not describe Francis as a saint in the heading: ‘Haec sunt verba sacrae admonitionis venerabilis patris nostri Francisci ad omnes fratres’.Footnote ⁹⁰ However, Esser omitted the word ‘sancti’ or ‘beati’ before ‘Francisci’, a point corrected in Grau's revised edition.Footnote ⁹¹ It seems, then, that the original argument can be maintained, namely that our manuscript dates from before Francis's canonisation in 1228. It should be dated within his lifetime or between his death and canonisation. If this is correct, it is the earliest extant manuscript of Francis's works with the possible exception of two celebrated autograph manuscripts, the original Regula bullata of 1223, and the manuscript of the Epistola ad clericos. All these, except the Regula bullata, contain very short works, in the case of the autographs the Laudes dei altissimi with the Benedictio fratris Leonis, and the letter to Leo.Footnote ⁹² They are considerably earlier than the next earliest manuscript, Assisi ms 338.

It is less easy to localise the manuscript than to date it. The script does not appear to give any clear pointers to where the manuscript was written. The five manuscripts drawn upon for comparison are probably all French or Italian, and it is likely that the new manuscript is French or Italian. The manuscript contains some distinctive spellings. It is worth considering these more fully, although insufficient work has been done on medieval Latin orthography to localise the manuscript on the basis of evidence of this type.Footnote ⁹³ The words ‘cognoscere’ and ‘cognitio’ appear as ‘congnoscere’ and ‘congnitio’, in the forms ‘congnovissetis’, ‘congnovistis’, ‘congnoscunt’ and ‘congnitionem’.Footnote ⁹⁴ The identical spellings occur in the text of the Admonitions as copied in Bodleian Library, Oxford, ms lat. theol.d.23 (beginning on fo. 139). This manuscript is known as the Little manuscript, after its former owner, the great Franciscan scholar A. G. Little.Footnote ⁹⁵ It is likely to date from the early fifteenth century and was written for Fr Lawrence of Rieti, minister of the province of St Anthony (i.e. Venice). Although the scribe was German,Footnote ⁹⁶ the manuscript was presumably written in northern Italy. The same spelling is adopted in Vatican Library, ms Rossin. lat. 23, dated by Esser and Oliger to the late thirteenth century but not localised by them.Footnote ⁹⁷ Another noteworthy spelling which occurs in this manuscript is x in place of s or ss: thus, ‘confexionem’ instead of ‘confessionem’ and ‘dexiderat’ instead of ‘desiderat’.Footnote ⁹⁸ This use of x has its parallel in another Italian manuscript, written in Verona in 1294. It has ‘Tomaxinus’ in place of the more usual spelling ‘Tomassinus’.Footnote ⁹⁹

The heading of the Admonitions, referring to Francis as ‘venerabilis patris nostri Francisci’, indicates that the manuscript was produced for members of the Franciscan order. While it is possible that a scribe, writing outside the order, or for the use of non-Franciscans, could have simply copied ‘nostri’ from an earlier exemplar, it is more likely that the manuscript was written in or for a Franciscan convent.

The very inconclusive palaeographical and orthographic evidence discussed above points to production in or for a Franciscan convent in France or Italy.

In the present manuscript, some of the Admonitions commence with a capital in red ink. In other cases, a space has been left for a capital but it has not been filled in. Thus, folio 5v displays five red capitals; there are five spaces where a capital has not been added, and in two of these there is an indication in the left margin of which letter was to be inserted. The explanation for this circumstance must be that the unexecuted capitals were to be in a colour other than red, most likely blue. The text scribe or another scribe added the red capitals but never got round to completing the others. Admonition 2 begins with the heading ‘Sermo’ and Admonitions 4 to 8 with the heading ‘Aliud’, all in red. These headings look like later additions, but there is no reason to suppose that they are in a different hand from that of the main scribe. The heading ‘Sermo’ doubtless reflects the origin of the second Admonition in a sermon or address by St Francis, perhaps delivered at a general chapter of the order.Footnote ¹⁰⁰

The newly recovered manuscript has the explicit ‘Hec sunt verba vite et salutis, quia si quis legerit et fecerit inveniet vitam et auriet salutem a Domino. Amen’, and two other thirteenth-century witnesses have a similar explicit.Footnote ¹⁰¹ It is worth noting that the first recension of the Epistola ad fideles, as it occurs in the Volterra manuscript, uses the identical words in its rubricated heading.Footnote ¹⁰²

The new manuscript has no marginal emendations or additions. In contrast, Assisi ms 338 was corrected thrice within the space of two lines, where the words ‘sanctis’, ‘ostendit’ and ‘in’ were inserted in the right and left margins towards the end of the first Admonition. Moreover, Assisi ms 338 has a marginal addition to the thirteenth Admonition in the form of a quotation from Matt. v. 9, which occurs in a minority of manuscripts but is absent from the manuscript on which this study is based. Here this Admonition begins with the words ‘[N]on potest congnoscere servus Dei quantam habeat patientiam’.Footnote ¹⁰³ This divergence may call to mind the statement in Jordan's Chronicle that Francis entrusted Caesarius of Speyer with ‘adorning’ the Rule with passages from the Gospels.Footnote ¹⁰⁴ Perhaps, then, the new manuscript transmits an earlier, ‘unadorned’ version of the thirteenth Admonition, although it must be admitted that there are numerous scriptural quotations elsewhere in the manuscript's text. The quotation is included in the edition by Esser and Grau but rejected by Paolazzi.Footnote ¹⁰⁵

The sequence of Admonitions in the newly available text differs from that of Assisi ms 338: the order of the eleventh and twelfth Admonitions is reversed. The other early manuscripts have the same sequence as the Assisi manuscript, and this sequence is followed in the critical editions by Esser and Grau and by Paolazzi.

Unlike the other manuscripts which may be dated to the thirteenth century, the present manuscript does not contain any headings for the Admonitions. It is likely that these were added to the text at some stage between the composition or compilation of the Admonitions and the copying of the Assisi manuscript. In this context, it is necessary to consider an important article by Carlo Paolazzi concerning the role of St Francis in the composition of the works ascribed to him.Footnote ¹⁰⁶ Paolazzi draws attention to the account in the Speculum perfectionis and the Compilatio Assisiensis of a visit by a doctor called Bonus Iohannes to Francis, who did not wish to call him by his name: ‘Non enim voluit eum vocare proprio nomine, quia nunquam volebat aliquem nominare qui vocaretur “bonus” propter reverentiam Domini qui dixit: “Nemo bonus nisi solus Deus”. Similiter nec “patrem” nec “magistrum” volebat aliquem vocare vel in litteris suis scribere, propter reverentiam Domini qui dixit: “Et patrem nolite vocare vobis super terram, nec vocemini magistri”, etc.’Footnote ¹⁰⁷ Examining the appearences of the words ‘bonus’, ‘magister’ and ‘pater’ in Francis's work, Paolazzi finds them used only to describe God or Christ. The one exception is the heading of Admonition 20, ‘De bono et vano religioso’. It is, however, generally recognised that the headings in the Admonitions do not derive from Francis himself.Footnote ¹⁰⁸ Paolazzi concludes that the wording of the Opuscula precisely reflects Francis's preoccupations; he convincingly rejects the view of Esser, among others, that Francis tended to dictate his works in the vernacular, which the scribe then translated into Latin. According to this view, even if the thoughts were those of Francis, the phraseology was not his.Footnote ¹⁰⁹ In the present context, it is worth reiterating that the headings do not occur in what appears to be the earliest manuscript of the Admonitions. In this respect, and in the omission of the word ‘beatus’ or ‘sanctus’ to describe the author, the manuscript undoubtedly represents an earlier stage in the transmission of the text than the other thirteenth-century manuscripts.

Since the headings are not part of Francis's original work, we should not be surprised to find variations in them from one manuscript to another.Footnote ¹¹⁰ This is apparent in the thirteenth-century manuscripts. The fourth Admonition, for instance, appears as ‘Ut nemo appropriet sibi prelationem’ in the Assisi manuscript, ‘Ut nullus sibi appropriet prelationem’ in the Florentine manuscript and as ‘Ut nullus sibi appropriat prelationem’ in the Volterra manuscript.Footnote ¹¹¹

The present manuscript and the early manuscripts at Assisi, Florence, London, Rome and Volterra all include the twenty-seventh Admonition, the authenticity of which has recently been the subject of debate between Carlo Paolazzi and Jan Hoeberichts.Footnote ¹¹² All the early manuscripts present this Admonition in continuous prose rather than the ‘verse’ form employed in the editions by Esser and Paolazzi,Footnote ¹¹³ a form reproduced in the latest translation of Francis's writings into English.Footnote ¹¹⁴ In this particular, Grau's revised edition of Esser is faithful to the early manuscripts.Footnote ¹¹⁵

The edition by Esser and Grau divides the manuscripts of the Admonitions into several groups and also distinguishes two individual manuscripts which do not fall into any of these groups.Footnote ¹¹⁶ The groups are as follows:

• Group a. This consists of eleven manuscripts. None of the distinctive readings of this group of manuscripts correspond to the text of the new manuscript.

• Group c. This group consist of a further eleven manuscripts. It represents an earlier stage in the transmission of the text than does Group a. None of the distinctive readings of Group c are found in the manuscript.

• Group d. Eleven manuscripts belong to this group, which is associated with Assisi ms 338 and the manuscripts closest to it, although these manuscripts form a separate group (see Group k below). None of the readings peculiar to Group d occur in the new manuscript.

• Group f. This group consists of sixteenth-century manuscripts and printed editions. As with Group d, there are affinities with Assisi ms 338. Only a very few of the distinctive readings of Group f occur in the manuscript.

• Group i. This consists of only two fourteenth-century manuscripts, which contain some major divergences from the other groups. None of these divergences are found in the new manuscript.

• Group k. This is the largest group, with at least fourteen witnesses, including Assisi ms 338 and all the earlier manuscripts of the Admonitions. The readings of the newly recovered manuscript tend to correspond to those of the manuscripts in this group.

While Esser's edition is generally regarded as a distinct advance on all previous editions, and the vast range of manuscripts consulted and evaluated is remarkable, it must also be admitted that it has been subjected to a considerable amount of criticism, especially in recent years. Stanislao da Campagnola, in the introduction to Fontes Franciscani, which reprints the edition by Esser and Grau, among other observations, criticises Esser for having selected the manuscripts which form the basis of the edition in an arbitrary manner and for giving only a selection of the variants in the apparatus criticus.Footnote ¹¹⁷ Ernesto Menestò argued that Esser failed securely to establish the interrelationship between the various manuscripts of Francis's Opuscula and thus to justify the readings adopted in the edition. He pointed out that the only stemma codicum in Esser's edition is the one for the Admonitions and that this is difficult to understand.Footnote ¹¹⁸ The shortcomings of Esser's edition are apparent in more concrete terms from a doctoral thesis by Laurent Gallant which provides an edition of the Officium passionis on the basis of a new study of the manuscripts.Footnote ¹¹⁹ They are also apparent in the criticisms of Esser's edition contained in three long articles by Carlo Paolazzi.Footnote ¹²⁰ He applies to the manuscripts of the Regula non bullata, of the Epistole and of the Admonitions the classical technique of textual criticism known as the Lachmann method in order to establish a stemma codicum and to prepare new editions of these works. Esser believed that Francis's Italianisms and errors or inelegancies in Latin were improved by the scribes to whom he dictated his works and by subsequent copyists. Esser for this reason tended to prefer readings which in his opinion were closer to Francis's original words, irrespective of which manuscript or manuscripts contained them. Paolazzi convincingly rejects such an approach.Footnote ¹²¹

Paolazzi's edition of the Admonitions is based on a new classification of the manuscripts. He argues that all the extant manuscripts derive from an archetype which contained two lacunae and one error.Footnote ¹²² Since the newly recovered manuscript displays the same defects, it doubtless also derives from the archetype. Paolazzi further distinguishes two branches, which diverged already in the thirteenth century, α and β.Footnote ¹²³

• Branch α consists of the following groups:

• Group α¹. This is in turn is divided into two subgroups, that is, Esser's groups a and c.

• Group f (in Esser's classification).

• Group i (in Esser's classification).

• Group k2. This contains ten manuscripts from Esser's group k, including the following thirteenth-century witnesses: Biblioteca Medicea Laurenziana, cod. X. Plut. XIX dextr.; Gray's Inn Library ms 2; Vatican Library, ms Ross. lat. 23; and Volterra, ms 225.

• Branch β consists of two groups:

• Group d (in Esser's classification).

• Group k1. This contains five manuscripts from Esser's group k, including Assisi ms 338 from the mid-thirteenth century.

The main differences introduced by Paolazzi in the classification of manuscripts of the Admonitions are the division of the manuscripts into two branches, and the separation into two of Esser's all-important Group k, some manuscripts being assigned to branch α (k2), others to branch β (k1); k2 and k1 represent the earliest stage of α and β respectively.Footnote ¹²⁴ Paolazzi provides a detailed discussion of the variant readings of the different branches and groups of manuscripts. It is worth considering how the text of the new manuscript relates to these.

Branch α. Paolazzi draws attention to fourteen passages in which α differs from the putative archetype.Footnote ¹²⁵ In nine of these, the readings of the new manuscript correspond to those of α; in the remaining five, they correspond to those of the archetype. Paolazzi then considers thirty-two passages in which the manuscripts of Group α¹ display variants from the archetype.Footnote ¹²⁶ In all but three cases, the newly recovered manuscript follows the archetype or at least is close to it; in those three cases, it differs from both the archetype and α¹. Next Paolazzi considers thirty-six passages in which the manuscripts of subgroup a differ from the archetype.Footnote ¹²⁷ In almost every case, the new manuscript is identical or close to the archetype. In one case it differs from both, and it never accords with the distinctive readings of Group a. There are twenty passages in which the manuscripts of subgroup c differ notably from the archetype.Footnote ¹²⁸ Here too our manuscript almost invariably coincides with the archetype. On just one occasion is its reading that of subgroup c. Similarly with the thirty-four divergent readings of Group f and the forty-two divergent readings of Group i, the new manuscript is identical or close to the archetype, except once when it differs from both the archetype and Group f and once when it has the same reading as Group i.Footnote ¹²⁹ However, the result of the comparison is different in the case of Group k2. Our manuscript contains seven of the ten distinctive readings of this group that Paolazzi lists.Footnote ¹³⁰

Branch β. Paolazzi considers seven divergent readings found in this branch, none of which are shared by the privately owned manuscript;Footnote ¹³¹ the latter in every instance coincides with the text as reconstructed by Paolazzi. Next Paolazzi turns to the distinctive readings of Group d.Footnote ¹³² In only two of the twenty-six passages that he discusses does the new manuscript give the reading of Group d; in the remaining cases it is identical to the edited text. Similarly, it shares the divergent readings of Group k1 only once (out of nine passages).Footnote ¹³³

One may conclude, on the basis of this comparison of our manuscript with the branches and groups of manuscripts according to Paolazzi's classification, that its readings coincide most closely with those of Group k1, to which all the other thirteenth-century manuscripts of the Admonitions, except Assisi ms 338, belong.

In the course of the discussion, we have mentioned various distinctive or notable readings in the newly recovered manuscript. We now wish to draw attention to some more of these readings, especially in comparison with Assisi ms 338 and the other thirteenth-century manuscripts, without, however, seeking to provide a comprehensive treatment of the variants.Footnote ¹³⁴

The title in the present manuscript reads: ‘Hec sunt verba sacre admonitionis venerabilis patris nostri Francisci ad omnes fratres. In nomine patris et filii et spiritus sancti. Amen. De corpore Domini nostri Iesu Christi.’Footnote ¹³⁵ The Assisi manuscript has a different order, commencing ‘In nomine patris et filii et spiritus sancti. Amen. Hec sunt verba sacre admonitionis venerabilis patris nostri sancti Francisci ad omnes fratres. De corpore Christi.’Footnote ¹³⁶ The Florentine manuscript differs from that of Assisi and reads: ‘In nomine domini. Amen. Hec sunt verba admonitionis venerabilis patris nostri beati Francisci.’Footnote ¹³⁷ The Volterra codex uses ‘sancti’, but like the Florentine, it omits ‘ad omnes fratres’.Footnote ¹³⁸ Gray's Inn Library ms 2 lacks the opening formula and the first Admonition, beginning with the words ‘Hec sunt verba Sancti Francisci ad fratres de ordine domini primo de malo proprie voluntatis’.Footnote ¹³⁹

• Admonition 1. The present manuscript opens with the words ‘Dixit Dominus Iesus discipulis suis’. Biblioteca Medicea Laurenziana, cod. X. Plut. XIX dextr. and four other manuscripts have ‘Dicit’ in place of ‘Dixit’. The same divergence occurs twice again in this Admonition. Esser and Grau prefer the present tense as ‘die ungewöhnlichere, aber dem Sprachgebrauch des hl. Franziskus entsprechendere Form’, but Paolazzi offers persuasive reasons for preferring ‘Dixit’.Footnote ¹⁴⁰ The sentence ‘Pater lucem h(abe)t inaccessibilem, et spiritus Deus’ (1.5) displays two divergent readings, neither of them acceptable. All other manuscripts have ‘habitat’ or ‘inhabitat’ in place of ‘habet’ and add ‘est’ before ‘Deus’. In the case of the clause ‘Ideo non eum spiritu videri potest’ (1.6), Esser and Grau prefer the reading ‘in spiritu’ as more ‘Italianate’, but Paolazzi argues for ‘spiritu’ alone.Footnote ¹⁴¹ The new manuscript is the only one to have ‘eum’ in this sentence, evidently an error for ‘nisi’, which occurs in the other manuscripts. This manuscript and a number of others, including those in the Biblioteca Medicea Laurenziana and Volterra, have ‘vivi’ before ‘dampnati sunt’ (1.8).Footnote ¹⁴² As in most other manuscripts, we find the reading ‘sanguis novi testamenti’ (1.10), although ‘sanguis meus novi testamenti’ seems preferable. Assisi ms 338 has the unsatisfactory ‘sanguis mei novi testamenti’.Footnote ¹⁴³ Like a small number of other manuscripts, our manuscript has a more extensive version of the question, ‘Ut quid non cognoscitis veritatem et creditis in Filium Dei’ (1.15), namely ‘Ut quod diligitis vanitatem et non cognoscitis veritatem et creditis in Filium Dei?’Footnote ¹⁴⁴ In the penultimate sentence of this Admonition, the manuscript gives ‘debet’ as a superscript addition before ‘esse’, although the word is in fact redundant at this point.

• Admonition 2. Here the manuscript has ‘dicit et operatur in ipso’ (2.3). Many manuscripts have ‘dedit’ instead of ‘dicit’, but the latter is the preferred reading in the editions by Esser and Grau and by Paolazzi. The following sentence has the unusual (although perfectly correct) spelling ‘subgestionem’ instead of ‘suggestionem’.

• Admonition 3. Our manuscript begins with the present tense (‘Dicit’), which is the preferred reading of Esser and Grau and of Paolazzi. The reading ‘perdet illam’ (at 3.2) is found in all but one of the thirteenth-century manuscripts and is adopted in the critical editions.Footnote ¹⁴⁵ The manuscript appears to be the only one which reads ‘Qui non renuntiaverit omnia’ rather than ‘omnibus’ (3.1). The text of Luke (xiv. 33) from which this is taken has ‘omnibus’. Paolazzi argues plausibly that the phrase ‘perdit corpus suum’ at 3.3, found in our manuscript and in all other manuscripts except those of Group a, should read ‘perdit [animam suam et] corpus suum’.Footnote ¹⁴⁶ Assisi ms 338, Volterra ms 225 and four other manuscripts have what Esser and Grau, followed by Paolazzi, regard as the earliest phrasing ‘sua voluntarie Deo sacrificet’ (3.5).Footnote ¹⁴⁷ Our manuscript, on the other hand, is among those which read ‘sua voluntate Deo sacrificet’. At 3.8, the manuscript, in a unique reading, gives the verbs ‘sustinuerint’ and ‘diligant’ in the plural rather than the singular. The singular would have been grammatically more correct, since the verbs' subject is ‘subditus’ in the previous sentence.

• Admonition 4. There is a wide variety of readings in the third and fourth sentences of this Admonition. Five thirteenth-century manuscripts (including the privately owned manuscript) and four later ones give ‘in abluendi fratrum pedes offitio deputati’. This is preferred by Esser and Grau as the lectio difficilior,Footnote ¹⁴⁸ and by Paolazzi. Our manuscript appears to be the only one to add a superfluous ‘esse’ at the end of the Admonition: ‘ad periculum anime esse ponunt’.

• Admonition 5. Here too the manuscript at one point provides an unsatisfactory text, not recorded elsewhere: ‘quia unus demon sit de celestibus et modo sit de terrenis’ (5.6). The critical editions read: ‘quia unus demon scivit de celestibus et modo scit de terrenis’.

• Admonition 6. At 6.2, the new manuscript reads ‘de hiis acceperunt a Domino vitam eternam’ against ‘de hi(i)s receperunt a Domino vitam sempiternam’ of the critical editions. No other manuscript displays the reading ‘acceperunt’, and ‘receperunt’ seems preferable.

• Admonition 7. At two points in this Admonition, the readings of branches α and β (in Paolazzi's classification) diverge; the newly recovered manuscript invariably coincides with α, confirming its association with Group k1 of branch α.Footnote ¹⁴⁹

• Admonition 8. A wide range of readings is evident at the beginning of this Admonition: ‘Ait apostolus’, ‘Ut ait apostolus’, ‘Sicut ait apostolus’, ‘Et ait apostolus’. Footnote ¹⁵⁰ The privately owned manuscript does not offer a clear solution here, for the first letter has been left blank for the insertion of a coloured initial. The second letter is t, which implies that the first word is ‘Ut’ or ‘Et’. The manuscript suggests that the variety of readings is the result of the use of an exemplar or exemplars in which the opening initial had not been filled in.

• Admonitions11 and 12. The order of these two Admonitions is reversed in the privately owned manuscript. The occurrence of ‘aliquam personam’ in place of ‘aliqua persona’ seems to be a simple grammatical mistake (11.2). Paolazzi argues for the reading ‘quia non remanet’ rather than ‘qui non retinet’ at 11.4.Footnote ¹⁵¹ This is supported by the present manuscript. The thirteenth-century manuscripts are divided over the words ‘ante oculos (suos)’ (12.3). The present manuscript includes ‘suos’, which is Paolazzi's preferred reading.Footnote ¹⁵²

• Admonition13. The phrase in the new manuscript ‘quantam ibi hinc(?) patientiam et humilitatem tantam habet et non plus’ (13.1) appears in the identical form in the critical edition by Esser and Grau, except that ‘hinc’ is omitted, while Paolazzi's text reads: ‘quantam ibi patientiam et humilitatem habet, tantam habet et non plus’. This phrase displays a large number of variations in the manuscripts.

• Admonition14. The last sentence of this Admonition begins ‘Hii non sunt spiritu’ in the new manuscript. It seems clear that ‘pauperes’ has been omitted in error before ‘spiritu’.

• Admonition 17. The new manuscript reads ‘quam non vult de se dare’ (17.2). The words are identical in Assisi ms 338 and certain other manuscripts, although the word order is different. An alternative is ‘quam vult’, but Esser and Grau prefer ‘quam non vult’ as the lectio difficilior, which is also Paolazzi's reading.Footnote ¹⁵³

• Admonition 18. In the second sentence of this Admonition, ‘Beatus servus, qui omnia bona reddit Domino Deo’, three thirteenth-century manuscripts, including the privately owned one, omit ‘sua’ after ‘bona’. This is undoubtedly the better reading, for as Esser and Grau and subsequently Paolazzi have pointed out, the use of ‘sua’ here would not accord with Francis's ideas concerning evangelical poverty.Footnote ¹⁵⁴

• Admonition 19. In the first sentence, the new manuscript has ‘quando magnificatus et exaltatus’ rather than ‘quando magnificatur et exaltatur’, which apparently occurs in all other manuscripts. The sense is identical in both readings. Later in the same sentence there is a spelling mistake, ‘simples’ instead of ‘simplex’. At 19.4, the manuscript has ‘subesse pedibus aliorum’, a reading preferred by Paolazzi to ‘esse sub pedibus aliorum’ adopted in the edition of Esser and Grau.Footnote ¹⁵⁵

• Admonition 20. In the last two sentences (‘et cum hiis producit homines ad amorem Dei in gaudio et letitia. Ve illi religioso qui delectat se in verbis otiosis et inanibus et cum hiis producit homines ad risum’), the new manuscript provides a series of readings favoured in the editions by Esser and Grau and by Paolazzi, with the exception of ‘inanibus’, to which Esser and Grau prefer ‘vanis’ as closer to the Italian.Footnote ¹⁵⁶

• Admonition 21. The first sentence of this Admonition appears with many variants in the manuscripts.Footnote ¹⁵⁷ The new manuscript has the unsatisfactory wording, ‘Beatus servus qui non loquitur que sub specie mercedis’, ‘que’ here being redundant. It may be that the scribe should have written, ‘Beatus servus qui, quando loquitur, sub specie mercedis’, this being the wording of Assisi ms 338. The manuscript shares the reading ‘parum fructum’ with most of the earlier manuscripts, in contrast to the alternative ‘parvum fructum’.Footnote ¹⁵⁸

• Admonition 23. Esser and Grau regard the word order ‘Fidelis et prudens servus’ as earlier than ‘Fidelis servus et prudens’ (23.3),Footnote ¹⁵⁹ but Paolazzi takes the opposite view, in which he is supported by the biblical text (Matt. xxiv. 46). The new manuscript has the former order, while Assisi ms 338 has the latter. Later in the same sentence, the new manuscript reads ‘interius punire per contritionem’ in common with Assisi ms 338, Biblioteca Medicea Laurenziana, cod. X. Plut. XIX dextr., and Volterra ms 225. Paolazzi regards ‘punire’ as a ‘diffrazione testuale’ from the original ‘penitere’ of Francis's text by the archetype (from which all the extant manuscripts derive).Footnote ¹⁶⁰

• Admonition 24. In the new manuscript one finds at the beginning: ‘Beatus servus qui tantum diligeret fratrem suum quando est infirmus quod non potest ei satisfacere quantum quando est sanus quod potest ei satisfecere’. Assisi ms 338 gives qui in place of the second quod and this is adopted in the critical editions of Esser and Grau and of Paolazzi. The reading of the new manuscript, which is shared by a large number of witnesses, echoes ‘quod non potest’ earlier in the sentence.

• Admonition 25. The new manuscript, in common with all but one of the thirteenth-century witnesses, reads: ‘non diceret de eo aliquid post ipsum’, while many other manuscripts omit ‘de eo’, followed by Esser and Grau. Paolazzi argues for the retention of ‘de eo’. Footnote ¹⁶¹

• Admonition 26. The new manuscript and most others begin ‘Beatus servus Dei’. Esser and Grau regard ‘Dei' as `eine spätere Angleichung an den Titel des Kapitels’,Footnote ¹⁶² the title in question being ‘Ut servi Dei honorent clericos’. The new manuscript contradicts this view, since it includes ‘Dei’ but has no chapter title. Paolazzi restores ‘Dei’, which is found in all the thirteenth-century manuscripts.Footnote ¹⁶³ The new manuscript appears to be unique in beginning the third sentence with ‘Ita’ rather than ‘Nam’.

• Admonition 28. The manuscripts offer three possibilities in the last sentence: ‘secreta Dei’, ‘secreta Domini’, or ‘secreta Domini Dei’. Esser and Grau regard ‘secreta Domini’ as the original reading,Footnote ¹⁶⁴ as does Paolazzi. The new manuscript is ambiguous, with ‘secreta D(omin)i’ or ‘secreta D(e)i’, while Assisi ms 338 has ‘secreta Domini Dei’.

The manuscript is relatively free from glaring errors. This is not to say that its readings should be automatically accepted as superior to those of other manuscripts. Indeed, some of these readings, as we have seen, are unsatisfactory. However, the new manuscript will undoubtedly play a prominent part in any further work on the text of the Admonitions. Some of its characteristics are already evident. It offers a virtually complete text of the Admonitions; no chapter is omitted. Nor does the text differ dramatically from that transmitted by the other thirteenth-century manuscripts. This shows that the Admonitions are likely to have attained their complete form at a relatively early date, that is, before Francis's canonisation (1228).

It is remarkable and fortunate that a number of objects survive which can be directly associated with the person of St Francis, not so much on the basis of pious tradition or wishful thinking as on the basis of reliable evidence: his autograph writings discussed above; the original letter of grace of Honorius iii of 1223 confirming the Franciscan Rule (the Regula bullata); the autograph writings of Francis' companion Leo, the earliest of which are thought to date from c. 1244;Footnote ¹⁶⁵ the breviary that Francis used;Footnote ¹⁶⁶ and the crucifix of S. Damiano that, according to the account in the Vita prima, spoke to Francis.Footnote ¹⁶⁷ There is a further group of objects where the connection to Francis is less close but which are notable for their early date, within Francis' lifetime or not long after his death. These include the manuscript of the Epistola ad clericos from Subiaco and, it seems, the fresco of Francis at Subiaco, both already mentioned. If the arguments advanced in this article are valid, to this second group should now be added the privately owned manuscript of Francis's Admonitions.