INTRODUCTION

             The Tractatus de Substantiis Separatis was written by St. Thomas Aquinas to his life-long friend and associate, frater Raynaldus or Reginaldus of Piperno. The fact that Raynaldus was entrusted with the editing of all the unfinished works of Aquinas, after the latter's untimely death in 1274, attests to the intimacy which obtained between the two men. There has never been any question concerning the authenticity of this treatise; it has always been characterized as "opus genuinum, absque dubio."

TITLE

             In the manuscripts which we have used for this edition of the text of the Tractatus de Substantiis Separatis, the title appears variously as Tractatus de Angelis, Tractatus de Substantiis Separatis, De Natura Angelorum, Libellus de Angelis sive de Substantiis Separatis, Liber de Angelis. In view of the fact that there does not seem to be any strong manuscriptural tradition for any one particular reading, we have decided to retain the title, Tractatus de Substantiis Separatis, by which the work has been traditionally known. Further, in no case have we found in the incipits of our manuscripts the rather time-honored but nevertheless nondescript name of "opusculum" attached to this particular work. As Eschmann has pointed out, "the notion of opusculum, i.e., a minor or smaller work, is in itself devoid of precise meaning (and) has in fact disserved the cause of Thomistic bibliography, if not even of Thomistic studies." While the Tractatus de Substantiis Separatis is shorter than some of the other so-called major works of the Angelic Doctor, it has nevertheless been characterized by Gilson as "an incomparably rich historical work" and by Eschmann as "one of the most important metaphysical writings of Aquinas." Henle has described the first chapter of this work as "the most elaborate synthesis of Platonic doctrine to be found in the whole Thomistic corpus." In order to preclude any possible classification of this work as one which belongs to "the miscellaneous matter of Aquinas' writings", we have decided to remain faithful to the testimony of our manuscripts and to incorporate the name Tractatus in the title of the work.

DATE

             The exact date of the composition of the Tractatus de Substantiis Separatis is unknown. Since the treatise is unfinished, (St. Thomas promises in Chapter XVII, no. 91 (p. 134), to treat of "quantum ad earum gubernationis ordinem"--which point is never developed), the presumption is that the work was composed just before the death of the Angelic Doctor. This is most likely the reason why Mandonnet and Glorieux were prompted to place its composition in the years 1272-1273. Walz gives a longer period, namely, 1261-1269. Callus dates the work as belonging to 1272. Grabmann, on the other hand, uses May 18, 1268 as the terminus a quo. From the colophon appearing in the majority of the manuscripts of William of Moerbeke's translation of Proclus' Elements of Theology, we know definitely that this was the day on which the translation was completed. St. Thomas quotes two passages from the Latin translation of the Elements in the Tractatus de Substantiis Separatis, i.e., Chapter XIX, nos. 110 and 113 (pp. 158, 161). Hence we are dealing with a work composed apparently in 1268 or later.

             The year 1270 has been offered as another possible terminus a quo. This is the date on which Book Kappa was first incorporated into the translation of the Metaphysics by William of Moerbeke. Book Lambda should therefore be quoted as Book XII in any work which was written after 1270. Eleven manuscripts used in the preparation of the present text of the Tractatus de Substantiis Separatis refer to Lambda as Book XII. It would therefore seem that 1270 is a terminus a quo for the dating of the composition of this work. Vansteenkiste, however, disagrees with this date and is inclined to place it in an earlier period, i.e., around 1259. Although Saffrey accepts the citation of Lambda as Book XII as a strong indication for dating a work as belonging to 1270 and after, Eschmann makes the valid observation that "since this work is unfinished and therefore not edited by Aquinas himself, its posthumous editor may perhaps be supposed to have interfered with details of this kind."

             Another possible corroboration for the year 1270 and even more specifically for December 10, 1270 as a terminus a quo is the striking passage in Chapter XIII, no. 67 (p. 108) of the Tractatus itself. In enumerating the positions of certain thinkers who erred not only as to the substance and order of spiritual substances but even as to their knowledge and providence, St. Thomas says, "Posuerunt Deum et alias substantias immateriales singularium cognitionem non habere nec inferiorum et praecipue humanorum actuum providentiam gerere."

             Now the order in which he mentions the specific points and the words themselves are almost a verbatim restatement of Propositions X, XI, and XII of the Condemnation of 1270 made by Bishop Etienne Tempier on December 10 against the masters at the University of Paris. The Propositions read: "X. Quod Deus non cognoscit singularia. XI. Quod Deus non cognoscit alia a se. XII. Quod humani actus non reguntur providentia Dei." If we bracket "et alias substantias immateriales" (which St. Thomas would naturally add in a treatise which dealt with the nature of separate substances), the similarity of the wording appears to be more than merely coincidental. The thirteen condemned theses were most certainly the object of much discussion by the teaching masters of the various faculties. One might say that they were on the lips of the whole university community, teachers and students alike. It would be most natural for St. Thomas, therefore, to incorporate the official wording of these theses and then proceed to a systematic refutation of them. By suppressing the names of the authors of these condemned positions, the Angelic Doctor was thus able to prescind from the atmosphere of heated controversy and to evaluate the propositions with calm impersonality and scientific objectivity. While we admit that this internal evidence is not absolutely conclusive per se, we nevertheless feel that it can be regarded as corroborative testimony in favor of late 1270 as a terminus a quo for the composition of the treatise. Further, the close relationship between the Tractatus de Substantiis Separatis and the Expositio super Librum de Causis as pointed out by Saffrey, would seem to indicate the period 1270-1272 as a very likely date for the composition of these two works. The establishment of a definitive date is still a matter for future chronologists of the works of St. Thomas Aquinas.

PLAN OF THE TRACTATUS DE SUBSTANTIIS SEPARATIS

Capitulum                                                                                                                           Divisio

I De opinionibus antiquorum et Platonis                                               1-7

II De opinione Aristotelis                                                                        8-14

III De convenientia positionum Aristotelis et Platonis            15-17

IV De differentia dictarum positionum Aristotelis et

                  Platonis                                                                                                18

V De opinione Avicebron et de rationibus ejus                       19-23

VI In quo positio Avicebron reprobatur                                                24-31

VII Quod spiritualis et corporalis substantiae non

                          potest esse una materia                                                                32-36

VIII De solutione rationum Avicebron                                                  37-45

IX De errore ponentium angelos non creatos et ejus

                  improbatione                                                                                       46-52

X Opinio Avicennae de fluxu rerum a primo principio

                          cum sua reprobatione                                                     53-59

XI De opinione Platonis de ideis et ejus improbatio   60-61

XII De errore ponentium omnes substantias spirituales

                  aequales esse creatas et ejus improbatio      62-66

XIII De errore dicentium Deum et angelos non habere

                  singularium notitiam                                                              67-76

XIV In quo ostenditur Dei providentiam se extendere

                  ad omnia                                                                                               77-79

XV Solutio positionum praedictarum                                                    80-85

XVI Error Manichaeorum circa praedicta et ejus

                          improbatio                                                                                    86-90

XVII Quid secundum fidem catholicam de origine

                          angelorum sit tenendum                                                              91-97

XVIII De conditione naturae substantiarum spiritualium

                          secundum fidem                                                 98-103

XIX De distinctione substantiarum spiritualium

                          secundum sacram doctrinam               104-114

             Although St. Thomas Aquinas had examined the nature of the angels under one or another aspect in a number of his works, the Tractatus de Substantiis Separatis is intended to be a veritable compendium of what every noteworthy writer, both pagan and Christian, had to say on the subject. In the opening phrases of his work, the Angelic Doctor states that his object in assembling the various opinions concerning the excellence of the angels is so that "we shall be in a position to accept whatever we find that agrees with faith and refute whatever is opposed to Catholic teaching."

THE TWO MAIN PARTS

             The Tractatus de Substantiis Separatis may be broadly divided into two main, although disproportionate, parts: (a) an examination of the opinions of the various philosophers on separate substances and a critique of each position strictly on philosophical grounds; (b) an exposition of Catholic teaching as found in Sacred Scripture, the Fathers, and especially Dionysius, "who excelled all others in teaching what pertains to spiritual substances." The first or philosophical part comprises Chapters I-XVI; the second part includes only Chapters XVII-XIX. This disproportion between the two divisions is to be explained only by the fact that St. Thomas Aquinas never completed the Tractatus de Substantiis Separatis. In Chapter XVII, no. 91 he states, "Since therefore it has been shown what the foremost philosophers, Plato and Aristotle, believed about spiritual substances as to their origin, the condition of their nature, their distinction and order of government . . . it remains to show what the teaching of the Christian religion holds about each individual point." Yet the treatise ends abruptly in Chapter XIX, no. 114, in which the third point under discussion, i.e., the distinction of spiritual substances, is left unfinished and the fourth point, i.e., the order of government, is completely omitted.

THE FIRST PART

             In the first or philosophical part, Chapter I deals with the opinions of the early Naturalists and Plato. Chapter II examines the position of Aristotle, who proceeded by a more manifest and surer way, namely, by motion, to investigate substances that are separate from matter. Chapters III and IV in turn, show the points of agreement and disagreement between Plato and the Stagirite. In Chapters V-VIII, St. Thomas explains and refutes the position of the author of the Fount of Life, who held that angels were composed of matter and form. In Chapter IX, he considers and rejects the opinion of those who said that separate substances were not created. Chapter X includes an examination and rejection of Avicenna's teaching of the procession of all things from the First Principle. Chapter XI criticizes the Platonists' position which asserts a certain order of causality among the spiritual substances. Chapter XII considers the error of Origen who held that all spiritual substances were created equal. In Chapters XIII-XV, St. Thomas studies and refutes the opinions of those who denied to God and to the angels a knowledge of singulars, as well as divine Providence to the First Principle. Chapter XVI repudiates the error of the Manicheans.

THE SECOND PART

             The second division comprises Chapters XVII-XIX. In the first chapter of the second division, St. Thomas sets forth what is to be held according to the Catholic faith concerning the origin of the angels. Like all other creatures, not only the highest but all spiritual substances were immediately produced by God. All the orders of these spiritual substances are established by divine disposition and not from the fact that one of them is caused by another. Further, they have goodness, being and life and other perfections from the same God Who is the essence of perfection. In Chapter XVIII, St. Thomas gives the Catholic teaching concerning the condition of the angels' nature. They are incorporeal and immaterial. They are not spirits united to bodies and they are in a place not in a corporeal but in a kind of spiritual manner. In the last chapter, viz., Chapter XIX, Aquinas begins but does not complete the distinction of the spiritual substances according to Sacred Teaching.

MANUSCRIPTS USED IN THE PREPARATION OF THE TEXT

I List of manuscripts:

A--CAMBRIDGE, Corpus Christi Libr. ms. 35          (fol. 131r-144r)

B--TOLEDO, Bibl. del Cabildo, Cod. 19-15 (fol. 86r-97v)

C--PARIS, Ste. Geneviève, Cod 238                             (fol. 88v-97r)

D--VENICE, S. Marco, Cod. 31, Cl. IV                       (fol. 263v-284r)

E--ROME, Cod. Vat. lat. 807                                                    (fol. 188v-215r)

F--ROME, Vat. lat. Ottob. 198                                     (fol 197v-207v)

G--PARIS, Bibl. Nat. lat. 14546                                   (fol 84r-100r)

H--BOLOGNA, Bibl. Univ. Cod. 861                                       (fol. 147r-156v)

I--BORDEAUX, Bibl. de la Ville, 131                         (fol. 204r-214v)

J--METZ, Bibl. de la Ville, 1158                                  (fol. 12v-13r)

K--PARIS, Bibl. Nat. lat. 15813                                   (fol. 180r-191r)

L--PARIS, Bibl. Nat. lat. 15814                                   (fol. 275r-290v)

II Description of manuscripts:

A--CAMBRIDGE, Corpus Christi Libr. ms. 35 (fol. 131r-144r)

Incipit: Quia sacris angelorum solemniis interesse non possumus . . . Explicit: . . . quos malos angelos dicimus ex inferiori ordine et corporeos esse. This is an early 14th century manuscript in which St. Thomas is referred to as "frater Thomas", although a later handwriting states, "in isto volumine continentur infrascripti libri Sancti Thome de Aquino". This manuscript is a member of the family of ABCD (i.e. TOLEDO, 19-15, PARIS Ste. Geneviève, 238, VENICE, 31, IV, and CAMBRIDGE, Corpus Christi, 35.) While these manuscripts definitely seem to possess a common parent, each has its own peculiarities and characteristics. CAMBRIDGE, Corpus Christi 35 contains chapter and paragraph divisions but no titles.

B--TOLEDO, Bibl. del Cabildo. Cod. 19-15 (fol. 86r-97v).

Incipit: Quia sacris angelorum solemniis interesse non possumus . . . Explicit: . . . quos malos angelos dicimus ex inferiori ordine et corpores esse. This manuscript belongs, according to Grabmann, to the second half of the 14th century. TOLEDO, 19-15 is the second member of the ABCD family of four manuscripts. Like CAMBRIDGE, Corpus Christi 35, it contains paragraph divisions but no titles.

C--PARIS, Ste. Geneviève, Cod 238 (fol. 88v-97r).

Incipit: Capitulum primum. De substantiis separatis ad fratrem Raynaldum. Quia sacris angelorum solemniis interesse non possumus . . . Explicit: . . quos malos angelos dicimus ex inferiori ordine et corporeos esse. Grabmann dates this manuscript just prior to the canonization in 1323. Castagnoli and Mandonnet place it as early 14th century and Mandonnet is even inclined to date is as very late 13th century.

PARIS, Ste. Geneviève, 238 in the third of the ABCD family indicating a common parentage. It contains chapter and paragraph divisions and is the only one of this particular group which has chapter titles.

D--VENICE, S. Marco, Cod. 31, Cl. IV (fol. 263v-284r).

Incipit: Quia sacris angelorum solemniis interesse non possumus . . . Explicit: . . . quos malos angelos dicimus ex inferiori ordine et corporeos esse. This manuscript, according to Grabmann, belongs to the 14th century, shortly after the canonization of the Angelic Doctor. It is the last of the ABCD family and contains chapter and paragraph divisions but no titles.

E--ROME, Cod. Vat. lat. 807 (fol. 188r-215r).

Incipit: Incipit tractatus de angelis Capitulum primum. Quia sacris angelorum solemniis interesse non possumus . . . Explicit: . . . quos malos angelos dicimus ex inferiori ordine et corporeos esse. This codex is part of a collection of St. Thomas' works which were prepared in 1317 for his canonization at the request of Pope John XXII. The manuscript is therefore dated as early 14th century. ROME, Vat. lat. 807 is very closely related to ROME, Vat. lat. Ottob. 198. They seem to have had a common source and their variants are consistently the same. This impression would seem to be at variance with Castagnoli's findings (at least for the Tractatus de Substantiis Separatis. Castagnoli had placed ROME, Vat. lat. 807 and ROME Vat. lat. Ottob. 198 (with certain reservations) in Group III (along with VENICE, 128, TOLEDO, 19-15, CAMBRIDGE, Corp. Christi, 35, and PARIS, Ste. Geneviève, 238). We have found, on the other hand, that ROME Vat. lat. 807 and ROME Vat. lat. Ottob. 198 seem to represent a different tradition from the others. Of the 12 manuscripts used in the preparation of this edition, these two evidence the strongest resemblance to each other, faithfully following even the same mistakes. In a few instances, ROME, Vat. lat. Ottob. 198 seems to be a superior manuscript, i.e., the chapters are numbered improperly in ROME, Vat. lat. 807 and in two chapter titles, the scribe insists on writing "articulorum" and, "articulis" instead of "Aristotelis".

F--ROME, Vat. lat. Ottob. 198 (fol. 197v-207v).

Incipit. Primum capitulum. Quid de angelis antiqui philosophi senserunt. Quia sacris angelorum solemniis interesse non possumus. . . Explicit: . . . quos malos angelos dicimus ex inferiori ordine et corporeos esse. According to Grabmann and Castagnoli, this manuscript belongs to the early 14th century. As pointed out above, there is a striking similarity between this manuscript and ROME, Vat. lat. 807 and we did not find the many inversions of which Castagnoli complains in his De forma absolutionis. The manuscript contains chapter and paragraph divisions and chapter titles.

G--PARIS, Bibl. Nat. lat. 14546 (St. Victor 635, fol. 84r-100r). Incipit: Incipit libellus de angelis sive substantiis separatis a fratre Thoma de Aquino editus. Capitulum primum. Quia sacris angelorum interesse solemniis non possumus . . . Explicit: . . . quos malos angelos dicimus ex inferiori ordine et corporeos esse. Mandonnet places this manuscript around the year 1280, i.e., before the end of the 13th century. Grabmann says that the handwriting belongs to the fourteenth century. Incidentally, he fails to list this tractate in his table of contents. It belongs between De entium quidditate (fol. 84r) and De intellectu (fol. 100r). We have selected this manuscript as basic, although it was necessary to accept a number of readings from other manuscripts and to make corrections and certain emendations. There does not seem to be any other manuscript in the group of twelve used which would indicate a parentage with PARIS, Bibl. Nat. lat. 14546 as e.g., in the case of ROME, Vat. lat. 807 and ROME, Vat. Ottob. 198 on the one hand, and CAMBRIDGE, Corpus Christi 35, TOLEDO, 19-15, PARIS, Ste. Geneviève 238 and VENICE, 31, IV, on the other. This manuscript is described in greater detail further on in the Introduction in connection with an examination of the basic text itself.

H--BOLOGNA, Bibl. Univ. Cod. 861 (1655) fol. 147r-156v).

Incipit: Incipit tractatus de substantiis separatis editus a sancto Thoma de Aquino ordinis fratrum praedicatorum. Prologus. Quia sacris angelorum solemniis interesse non possumus . . . Explicit: . . . quos malos angelos dicimus ex inferiori ordine et corporeos esse. Grabmann dates this codex as belonging to the early 14th century. Castagnoli places this manuscript along with METZ, Bibl. de la Ville, 1158 (which appears in our edition as "J") in Group I. It is a substantially good manuscript but PARIS, Bibl. Nat. lat. 14546 is definitely superior. The Bologna manuscript contains chapter and paragraph divisions and titles of chapters which are written by a different hand.

I--BORDEAUX, Bibl. de la Ville, 131 (fol. 204r-214v).

Incipit: Incipit tractatus Sancti Thomae de substantiis separatis. Quia sacris angelorum solemniis interesse non possumus. Explicit: . . . quos malos angelos dicimus ex inferno ordine et corporeos esse. The handwriting would seem to place this manuscript as belonging to the 14th century, around the time of the canonization. BORDEAUX, 131 seems to have a certain common parentage with the ABCD group, judging by the numerous variants they have in common. An examination of the apparatus reveals the great number of times the same variant is shared by the group ABCDI. This manuscript, however, is positively the worst of the twelve used for the edition of our text. It is replete with omissions, unintelligible additions, homoioteleuta, and incomprehensible passages. Pauson's observation "the scribe of BORDEAUX 131 was, however, most inaccurate, inverting, omitting and changing almost at will" is only too correct. On the other hand, curiously enough, in Chapter XIX, no. 110 (p. 158), BORDEAUX 131 is the only manuscript which contains a correct reading of Liber Divinarum Elementationum; all the other 10 manuscripts are incorrect. BORDEAUX 131 contains chapter and paragraph divisions. The chapter titles, however, were added by another hand. There are also written in the margins numerous summaries of the subject matter.

J--METZ, Bibl. de la Ville, 1158 (fol. 12v, 13r).

Incipit: Liber de natura angelorum. VI. Quia sacris angelorum interesse solemniis non possumus . . . This manuscript, which originally belonged to the library of the Abbaye du Parc near Louvain, was presented to the Bibliothèque de la Ville de Metz in 1880 by a collector named Salis. Mandonnet and O'Rahilly date METZ 1158 as belonging to the late 13th century; Kruitwagen places it in the year 1282. Castagnoli, in the Introduction to his edition of the De forma absolutionis, and Rossi in his Expositio salutationis angelicae pointed to the antiquity and outstanding worth of this manuscript. Weighing these opinions, we saw in METZ 1158 a strong possibility for a basic manuscript for our edition of this text and took steps to acquire a copy of it on microfilm.

             J. Pauson, in his St. Thomas Aquinas, De principiis naturae, reports: "According to the present librarian of the Bibliothèque de la Ville de Metz, during the recent war, Codex 1158 was deposited in a fortification and presumably destroyed by fire. Fortunately, however, Dom Lottin, O.S.B. of the Abbaye de Mont César at Louvain, had microfilmed this manuscript about 20 years previously and the reproduction is still in the possession of the library of the Abbaye."

             Professor E. Gilson, in communicating with Dom O. Lottin about the availability of this manuscript of the Tractatus de Substantiis Separatis, was informed that unfortunately Dom Lottin had photographed only folios 4-13 on microfilm. As he states in his letter to Professor Gilson, "Le manuscrit 1158 de Metz est bel et bien detruit. Et j'en ai une partie en microfilm. Malheureusement, je ne sais quel mauvais génie a guidé mon photographe: je n'ai que le debut du folio 4 a 13." We have, therefore, only folios 12v and 13r which extend from Chapter I, no. 1 to Chapter II, no. 13 (pp. 35-49).

             The variants which occur in this short section of the work have been included in the apparatus as "J" readings. It is especially interesting to note that of the 106 variants which we have found in METZ 1158, 73 are contained in none of the other eleven manuscripts used. While obviously the brevity of the available text precludes any possibility of a definitive classification, it would seem that METZ 1158 is representative of a substantially different manuscriptural tradition not represented by any of the others used in the preparation of our text. It is also interesting to note that METZ 1158 and BOLOGNA 861 which Castagnoli grouped together in Class I in his De forma absolutionis, do not seem to indicate such close affinity to the De Substantiis Separatis. We must remember, however, that we have an extremely truncated text of METZ 1158. It is truly regrettable that such an important manuscript perished, thereby creating a probably irreducible lacuna in the eventual establishment of a truly critical text of the Tractatus de Substantiis Separatis. METZ 1158 contains chapter divisions but no titles and no paragraph divisions.

K--PARIS, Bibl. Nat. lat. 15813 (fol. 180r-191r).

Incipit: Tractatus de angelis editus a fratre Thoma de Aquino ordinis praedicatorum. Quia sacris interesse angelorum solemniis non possumus . . . Explicit: . . . quos malos angelos dicimus ex inferiori ordine et corporeos esse. Perrier used this manuscript in the preparation of his text of the De substantiis separatis but he neither describes it nor makes any attempt to date it. The handwriting of PARIS, Bibl. Nat, lat. 15813 would seem to indicate that we are dealing with a late 13th or early 14th century manuscript. St. Thomas is referred to as frater rather than sanctus. The manuscript contains chapter and paragraph divisions but no titles. Both PARIS, Bibl. Nat. lat. 15813 and PARIS, Bibl. Nat. lat. 15814 seem to have a common source.

L--PARIS, Bibl. Nat. lat. 15814 (fol. 275r-290v).

Incipit: Incipit liber de angelis editus a fratre Thoma de Aquino ordine fratrum praedicatorum. Capitulum primum. Quid de angelis antiqui philosophi senserunt, Quia sacris angelorum interesse solemniis non possumus. . . Explicit: . . . quos malos angelos dicimus ex inferiori ordine et corporeos esse. Because this manuscript, like "K", refers to St. Thomas as frater rather than sanctus, we are inclined to place it in the early 14th century, prior to the canonization of the Angelic Doctor by Pope John XXII. PARIS Bibl. Nat. lat. 15814 contains chapter and paragraph divisions and titles. Corrections appear in the margins and short summaries of the doctrine contained in the text are made in a more cursive hand than that of the body of the manuscript.

III Classification of manuscripts

             The only attempt to classify the numerous manuscripts of the shorter treatises by St. Thomas Aquinas was made by Castagnoli in his edition of the De forma absolutionis. He divides his 31 manuscripts into 4 groups of which the following were chosen as the best and most representative:

I           METZ 1158                                                                     (14th cent.)

             BOLOGNA 861                                                               (14th cent.)

II          PARIS 16297                                               (1270-1273)

               (containing only five treatises)

III         VENICE (S. Marco) 128                                   (14th cent.)

             TOLEDO 19-15                                                              (14th cent.)

             CAMBRIDGE, Corpus Christi 35       13th-14th cent.)

             ROME, Vat. lat. 807                                          (14th cent.)

             PARIS, Ste. Geneviève 238                   (14th cent.)

             ROME, Vat. lat. Ottob. 198                 (14th cent.)

IV        MUNICH 3754                                           (14th-15th cent.)

             BRUSSELS 1573                                                            (15th cent.)

---         PARIS, Bibl. Nat. lat. 14546                13th cent.)

             Castagnoli considers the last two groups inferior to Groups I and II. Group II, he holds, is more important but inferior to Group I, which is better and closer to the original. (PARIS 16297 which represents Group II contains only five treatises.) PARIS Bibl. Nat. lat. 14546 (13th cent.) is placed in a class all its own. The text of this manuscript according to Castagnoli, is that of Godfrey of Fontaines with marginal corrections made directly on the autograph.

             Rossi does not classify the manuscripts which he used for the text of Expositio salutationis angelicae. He does indicate, however, that METZ 1158, PARIS, Nat. lat. 14546, and POMMERS-FELDEN 90-2656 are the best of the 19 manuscripts which he used. Keeler indicates in his Tractatus de unitate intellectus, which is based on 20 manuscripts, that PARIS, Nat. lat. 14546 is a substantially good manuscript. Finally, Perrier made this manuscript the basis for the corrected text of the treatise which he edited in his Opuscula omnia necnon opera minora.

PREPARATION OF THE TEXT

             To our knowledge, the only serious attempt to present a corrected text based on earlier manuscripts of the Tractatus de Substantiis Separatis was made by J. Perrier in 1949. (A more recent edition by R. Spiazzi, Rome, 1954 is only a corrected text of the printed Piana and Parma editions. Unfortunately, Perrier's text represents a very limited manuscriptural tradition, since its editor used but four manuscripts and all of them Parisian. Perrier's edition is based substantially on PARIS, Nat. lat. 14546 which, as we have seen from the testimony above, has been judged to be a reliable and even superior text. The remaining three manuscripts used, were (1) PARIS, Ste. Geneviève 238, (2) Bibl. Nat. lat. 15813 and (3) Bibl. Nat. lat. 15814. Finally variants were included in the apparatus from the printed Morelles edition of 1612, which is itself a reprint of the earlier 1570 (Piana) edition.

             In the preparation of our text, we were faced with the problem of selecting the best manuscript to serve as basic. To single out one manuscript from a fairly large number of available ones and to make it the definitive text is very difficult. As Vansteenkiste so correctly points out, the problem resolves itself into the question, whether it is possible to give a good text by using a single manuscript. Actually, "this is a question of words. If we use a second manuscript, we shall have to say that we used two manuscripts as basic. The case when one can use almost exclusively a single manuscript as basic is quite exceptional."

             With this in mind, we began testing the various manuscripts at our disposal. We found that no single manuscript was perfect; every one had imperfections and shortcomings, whether by way of omissions, homoioteleuta, or outright errors. Two manuscripts, however, stood out as being better than the others, namely, METZ de la Ville 1158 and PARIS, Bibl. Nat. lat. 14546. Dom Lottin's communication concerning the unavailability of a copy of the destroyed METZ manuscript, eliminated this as a possible basic. Hence, PARIS, Bibl. Nat. lat. 14546 was selected as basic but with certain reservations. Because of the number of unsatisfactory readings contained in Perrier's text, based on PARIS Bibl. Nat. lat. 14546, we also selected 10 other manuscripts which would furnish us a wide variant reading from other traditions. The 10 manuscripts chosen are, to our mind, adequately representative of Castagnoli's first three Groups, described above, which he considered superior to the other manuscripts. Group I is represented in our text by BOLOGNA 861. (Group II is discounted because PARIS 16297 contains only five treatises.) Group III is represented by VENICE 31, IV; TOLEDO 19-15; ROME, Vat. lat. 807; CAMBRIDGE, Corpus Christi 35; PARIS, Ste. Geneviève 238 and ROME, Vat. lat. Ottob. 198. In addition, we have also used PARIS, Bibl. Nat. lat. 15813, PARIS, Bibl. Nat. lat. 15814 and BORDEAUX, Bibl. de la Ville 131. In this way, we feel that the resulting text, even if far from a critical one, should be more satisfactory than that of Perrier because it is more representative of the extant manuscripts of this treatise. It will remain for the Leonine scholars to give us a definitive critical text of the Tractatus de Substantiis Separatis. Meanwhile, we believe, a readable and useful text has been made available to the serious student of St. Thomas Aquinas. To achieve such a text was our goal.

VARIANTS

             In a number of instances, we have rejected the readings of our basic manuscript, PARIS, Bibl. Nat. lat. 14546. These are noted in the apparatus. As the reader will observe, we have also incorporated the majority of suggestions made by C. Vansteenkiste in his review of Perrier's text.

             The first class of emendations concerns the chapter divisions and titles. After considerable evaluation and comparison with other manuscripts, we have decided to abandon altogether the chapter divisions and notations of PARIS, Bibl. Nat. lat. 14546. These chapter headings which appear for the most part in the margin of the text were added by a different and presumably later hand. The following instance will indicate our reason for rejecting the divisions and readings of our basic manuscript. Chapter I of PARIS, Bibl. Nat. lat. 14546 contains no title whatsoever. Chapter II is marked by the later scribe in the margin opposite Chapter I, no. 2 (p. 36) beginning with "Sed Epicurei ex Democriti doctrinis originem sumentes . . ." It reads as follows: "Capitulum secundum. Unde habuit originem secta Epicuriorum". Curiously, only two other manuscripts take up this division and heading. ROME, Vat. lat. 807 incorporates the title into the text preceding the chapter division, i.e., after "incorporeos esse videbatur quos Angelos nominamus". Further, it marks this new chapter "Capitulum primum" even though the following chapter (no. 15, p. 51) is titled "Capitulum tertium". The other manuscript which belongs to the same tradition as ROME, Vat. lat. 807 is ROME, Vat. lat. Ottob. 198 as was indicated above. It likewise gives the same heading, i.e., "Unde habuit originem secta Epicuriorum" but does not make it a new chapter and consequently the following chapter i.e., no. 15 appears as Chapter II. Hence only PARIS, Bibl. Nat. lat. 14546 and ROME, Vat. lat. 807. designate no. 15 as Chapter II and end up with one number higher than the other manuscripts. We therefore rejected this division in our basic and have followed the tradition of the other manuscripts.

             The second class of emendations concerns those made in the body of the text itself. Although we have followed our basic manuscript substantially, the reader will observe a considerable number of changes. These variant readings have been taken for the most part from other manuscriptural traditions as indicated in the apparatus. In some few instances, in which not one of the 12 manuscripts used for our edition contained the correct reading, an emendation was made. Such was the case in Chapter II, no. 11, 1. 9, (p. 46). The passage reads, "Tentaverunt enim quidam sectatorum Aristotelis, ut patet in Epistola Porphyrii ad ANEBONTEM Aegyptium . . ." Although St. Thomas mentions the title of this epistle again in Chapter XIX, no. 106, (p. 152), no manuscript contains the correct reading in Chapter II, no. 11; hence the emendation.

LITERARY PROBLEMS

             In the preparation of a critical or even a corrected text of a given work, the editor first examines the extant manuscripts at his disposal. If the number of manuscripts is so large that it proves to be unwieldy, he will try to collate and classify them according to certain traditions and, when possible, to trace them to a common stemma. He will select the manuscript which appears to be best and this he makes his basic. Almost invariably, however, he will find it necessary to introduce readings from other manuscripts to substitute for errors, omissions, homoioteleuta and the like. When the number of extant manuscripts is so great that there is a veritable plethora of them (as in the case of the so-called "opuscula" of St. Thomas Aquinas which appear as members of large fixed collections and also as individual works which have been copied and re-copied a great number of times), the problem is more complicated. The editor of a corrected text must make certain judicious choices and content himself with a substantially correct and readable text based on the manuscripts which he has selected or which have been available to him. The editors of a genuinely critical text of the short treatises of St. Thomas have before them a formidable task. It will be their duty to locate, examine, and evaluate each extant manuscript of the particular work, whether it be a member of the large collections or an independent copy of a single treatise. The truly critical text will therefore represent the best readings culled from all the extant manuscripts and such a work will be final and definitive.

             Concerning the relative value of the manuscripts used for any particular edition, the general presumption is that the earlier a manuscript is, the more correct it is because it is closer to the autograph and hence fewer corruptions have crept into the text. On the basis of this reasoning, the incunabula or early printed editions have no place in the preparation of a critical or even corrected text. In many cases, these printed editions contain a great number of errors and it is precisely the corrupt state of these printed works replete with inaccuracies, which makes the edition of a critical or corrected text mandatory. Usually, then, the only reason these printed editions have been incorporated into the critical apparatus has been to show how erroneous their readings have been in comparison with the correct readings found in the manuscripts.

             The situation, however, does not seem to be quite so clearcut. The problem becomes a little more complicated when we consider the question of the manuscripts which were used in the printing of the incunabula. Granted that the printed edition is chronologically a later work than any of the extant manuscripts of a particular treatise, the pivotal question seems to be the following: What is the value of the manuscript or manuscripts which the editor used for the printed edition? If the manuscript is still extant, an evaluation can be made and an answer readily given; this offers no particular problem. But if the manuscript is no longer available, the situation becomes more confused. For example, what proof do we have that the editor of the printed edition did not have a much earlier and far superior manuscript at his disposal than any of those which we have today? In reality, the extant manuscripts would therefore be inferior to the text of the printed edition. Barring certain errors which the printer of the early incunabulum might have made, could not this printed edition, which is a reasonably faithful representation of the early manuscript, have as much value in the establishment of a critical text as the manuscript itself, were it extant? Thus, does not such an incunabulum as the Soncinas or the Pizzamanus for the so-called "opuscula" of St. Thomas Aquinas merit a place "a pari" with the manuscripts?

             This particular aspect of the whole question concerning the preparation of critical editions is treated very pointedly by P.

O'Reilly in the Bulletin thomiste in his review of J. Pauson's edition of St. Thomas' De Principiis Naturae. Introduction and Critical Text. As O'Reilly points out in his examination of Pauson's text, there is nothing to prevent us from holding that Soncinas had at his disposal all the oldest and best manuscripts, and possibly even the autograph itself. O'Reilly contends that many of the Soncinas readings are superior to those in Pauson's critical text. Again, M. Martin, in his Notes critiques au sujet de l'Opuscule IX de saint Thomas d'Aquin, ses manuscrits, ses éditions, found that the only three known manuscripts of this work were very corrupt and inferior to the Piana printed text based on a better but also lost manuscript. We therefore seem to have a situation in which a printed text (since it represents a lost manuscript superior to the extant ones) should be used in the preparation of a critical text and possibly even be made a basic text for the edition.

             In our own work on the Tractatus de Substantiis Separatis, we found an interesting point which might be added to the above thesis concerning the value of printed editions. The 12 manuscripts that we have used in the preparation of the present text are sufficiently representative of the basic classifications which Castagnoli had made. We therefore feel reasonably certain that we have a fairly accurate and readable text.

             In Chapter I, no. 2 (p. 35), we found a reading which, as it stands, does not seem to be acceptable and therefore calls for some emendation. The basic difficulty, however, was the fact that we had little authority for making this emendation. Consider the text itself. St. Thomas is giving a short resumé of the philosophical positions of the early Greek Naturalists. He says, "The first of those who philosophized on the nature of things, believed that only bodies existed and held that the first principles of things were certain corporeal elements, either one or several." The text then continues: "Et si unum, aut aquam ut Thales Milesius, aut aërem ut Diogenes, aut ignem ut Hippasus, aut vaporem ut Heraclitus." Now the name of Heraclitus is regularly associated with fire. No sources of Greek philosophy ever attribute "vapor" to Heraclitus. It would seem therefore that "ut Hippasus aut vaporem" is an interpolation. The disconcerting point, however, is that eleven of the twelve manuscripts used in the preparation of our text attribute "vapor" to Heraclitus. (Manuscripts A and B, i.e., CAMBRIDGE, Corpus Christi Libr. 35 and TOLEDO, Bibl. del Cabildo 19-15 do omit "Hippasus" but still attribute both fire and "vapor" to Heraclitus--"aut ignem aut vaporem ut Heraclitus"). Manuscript D, i.e., VENICE S. Marco 31, IV omits "ut Hippasus aut vaporem ut" and substitutes "ac" for the entire phrase, leaving, at the same time, a space between "ac" and "Heraclitus". The presumption may be safely made that the scribe of VENICE, S. Marco 31, IV deleted the troublesome phrase, leaving a space wherein he could insert a later correction. Hence "D" reads, "aut aërem ut Diogenes aut ignem ac-----------Heraclitus." All twelve manuscripts therefore seem to be representative of a tradition which contains this interpolation. An examination of the 1488 Soncinas edition and the 1490, 1498 and 1508 Pizzamanus editions reveals a very startling reading. All four incunabula give the following reading: "Aut aërem ut Diogenes aut ignem ut Heraclitus". It would seem to us that the incunabula are correct and that the manuscripts contain interpolations.

             Similarly, in Chapter XIX, no. 106 (p. 153), all manuscripts read "essent" in the following passage: "Unde impossibile est in daemonibus inveniri naturalem inclinationem ad malum, si ESSENT pure intellectuales non habentes admixtionem naturae corporeae." Now clearly SUNT would seem to be a much better reading but none of the manuscripts warrants it. A check with the 1488 Soncinas as well as the 1490, 1498 and 1508 Pizzamanus printed editions revealed that all four incunabula read SUNT; hence once again we settled in favor of the incunabula.

             The only logical conclusion one can draw is that all four printed editions--or, at least, the Soncinas if the Pizzamanus editions substantially reproduce the Soncinas text--were based on some manuscript or manuscripts that we no longer possess. Neither can it be argued that Soncinas did not base his text on an entirely different and presumably older but no longer extant manuscript, but merely made appropriate emendations as he was setting up type for the printed editions. An examination of the first 13 sections of the Soncinas edition of the De substantiis Separatis, i.e., Chapter I no. 1--Chapter II, no. 13 (pp. 35-49) revealed 99 variant readings not found in our established text. Of the 99 variants, 50 are not found in ANY of the twelve manuscripts used in the preparation of the text. Further, the use in the Soncinas text of such words as e contrario instead of e converso, which latter phrase is found in all 12 manuscripts (Chapter I, no. 12 p. 48); of materialem for the 12 readings of corporalem (Chapter I, no. 3 p. 37); of primo for supremo (Chapter II, no. 10 p. 45) would definitely indicate that a manuscript of an altogether different and, in some cases, more correct tradition was used as the basis for the printed edition. Hence the "printer's correction theory" must be rejected.

             It is true that our conclusion must be somewhat qualified, since we have only twelve manuscripts at our disposal. Two of these manuscripts, however, namely, METZ, Bibl. de la Ville 1158 and PARIS, Bib. Nat. lat. 14546, have been judged by several editors of previous Thomistic works, as we noted above, to be superior if not the best of all manuscripts extant. This has also been our finding, at least for PARIS, Bibl. Nat. lat. 14546, since there is only one folio, i.e., 12v and 13r in existence of METZ, Bibl. de la Ville 1158 of the Tractatus de substantiis separatis. In line with the findings of O'Reilly and Martin, it seems appropriate to suggest that greater importance should be attached to the early printed editions in the preparation of a critical text. Once an incunabulum has been ascertained to be truly representative of a good and presumably early manuscript no longer in existence today, it would seem that the value to be attached to it, with due allowance made for printers' mistakes, is that of the lost manuscript it represents. Further research along these lines should prove to be extremely fruitful.

Abbreviations

LIST OF ABBREVIATIONS

AACTA -- K. Foster and S. Humphries, Aristotle's De Anima in the Version of William of Moerbeke and the Commentary of Saint Thomas Aquinas, New Haven, 1951.

BW -- A. C. Pegis, Basic Writings of Saint Thomas Aquinas, 2 vols. New York, 1945.

CP -- E. Gilson, The Christian Philosophy of Saint Thomas Aquinas, New York, 1956.

Elem -- Proclus, The Elements of Theology, ed. E. R. Dodds, Oxford, 1933.

HCP -- E. Gilson, History of Christian Philosophy in the Middle Ages, New York, 1955.

OBE -- St. Thomas Aquinas, On Being and Essence, tr. A. Maurer, Toronto, 1949.

OSC -- ----------, On Spiritual Creatures, tr. M. C. Fitzpatrick, Milwaukee, 1949.

OTCF -- ---------, On the Truth of the Catholic Faith, 5 vols., tr. A. C. Pegis, J. Anderson, V. Bourke, C. O'Neil, New York, 1957.

PG -- J. P. Migne, Patrologiae Cursus Completus, Series Graeca, 162 vols., Paris, 1857-1866.

PL -- -----------, Patrologiae Cursus Completus, Series Latina, 221 vols., Paris, 1844-1864.

SCG -- St. Thomas Aquinas, Summa Contra Gentiles, ed. Leonine, Rome, 1934.

ST -- -----------, Summa Theologiae, ed. Leonine, vols.

4 - 12, Rome, 1882-1948.

  -----------------, ed. Ottawa, 5 vols. 1941-1945.