kommentierung, Überlieferung, nachleben () || boethius as an aristotelian scholar

STEN EBBESEN, Kopenhagen

BOETHIUS AS A N ARISTOTELIAN SCHOLAR*

By the end of the fifth century A. D. Greek philosophy had for cen-turies been dominated by Aristotelo-Platonic scholasticism.1 The scholastic attitude is well expressed by Ammonius who says that it is with good reason that earlier philosophers have established the tra-dition of introducing the reading of an authoritative text with an inquiry into its aim (σκοπός), utility and authenticity; for (1) if you do not know the aim, you are likely to drop the book before reach-ing the end of it; (2) if you do not know what it is useful for, you are not likely to embark with enthusiasm on the reading; and (3) even after having been told what it is good for, "we are apt to doubt the utility of a book until we are assured that it is a genuine work of a classical author known to be generally esteemed, like Aristotle and Plato-for as far as they are concerned, we assume that whatever they have said is useful."2

* Editions of Boethius' works. Inst. Arithm.: Anicii Manlii Torquati Severini Boetii De institutione arithmetica . . . ed. G. Friedlein, Lpz. 1867. In Intr. 1 a ( = ed. prima) & 2a: Anicii Manlii Severini Boethii In Isagogen Porphyrii commenta . . . rec. S.Brandt ( = CSEL 48), Wien-Lpz. 1906. In Cat.: Patrologia Latina (Migne) 64. In Int.: Anicii Manlii Severini Boetii commentarii in librum Aristotelis ΠΕΡΙ ΕΡΜΗΝΕΙΑΣ rec. C.Meiser, Lpz. 1877-80 (1 a = ed. prima in vol. 1; 2a in vol.2).

Introductio ad Syllogismos categoricos, De syllogismo categorico, De divisione, In Topica Ciceronis, De differentiis topicis: Patrologia Latina 64. Hyp. Syll.: A. M. Severino Boezio, De hypotheticis syllogismis, ed. L. Obertello ( = Istituto di filosofia dell' Universita di Parma, Logicalia 1), Brescia 1969. Translations of Porphyry and Aristotle: Aristoteles Latinus 1-6. As a supplement to Migne's edi-tions of In Cat. and Syll. Cat. I have used ms. Thott 166-168, 2°, Royal Library, Copenhagen (tenth century).

1 On ancient scholasticism, see S. Ebbesen, Commentators and Commentaries on Aristotle's Sophistici Elenchi, Leiden 1981,1,52 sqq. ( = Corpus Latinum Commen-tariorum in Aristotelem Graecorum VII 1); id., Ancient Scholastic Logic as the Source of Medieval Scholastic Logic, in: The Cambridge History of Later Medie-val Philosophy, ed. N. Kretzmann, A. Kenny, J. Pinborg, Cambridge 1982, 101 sqq.

2 Ammon., In Intr. 21.

Brought to you by | New York University Bobst Library Technical ServicesAuthenticated

Download Date | 12/7/14 11:27 PM

Boethius as an Aristotelian scholar 287

In periods when the scholastic attitude to the auctores prevails, the exegesis of the classics of philosophy acquires great importance. By the year 500 a considerable number of commentaries and auxil-iary treatises relating to the works of Aristotle and Plato had been written in Greek. Quite a few would be available to anyone with the wish and the financial means to obtain them. But in spite of an attempt in the fourth century to produce a Latin scholastic library,3

philosophy had not come to Latium. It had to wait for Boethius. How little his predecessors had achieved may be gauged from the fact that he has no standard Latin equivalents for several elementary Greek terms. To render όργανον he feels obliged to use a hendi-adys, ferramentum et quodammodo supellex* For φαντασία he uses visum. The term had been coined by Cicero five hundred years ear-lier, but Boethius introduces it in a way which shows that he ex-pects his readers to be as unfamiliar with it as were Cicero's con-temporaries.5 To benefit from Plato's and Aristotle's useful writings a mastery of Greek was still required.

Boethius came to see it as his mission to bring Latin philosoph-ical literature up to contemporary Greek standard. After an initial attempt to build on the foundations laid in the fourth century,6 he decided he had to start from scratch, translating the auctores and then adding the necessary auxiliary works. At its most ambitious, his plan seems to have comprised:7

3 Texts relating to the Organon: Isagoge (free translation) + commentary + mono-graph on definition, all by Victorinus; paraphrase of Cat. ( = Ps.-Augustine, Cate-goriae Decern); paraphrases of Anal. Pr. & Anal. Post, by Themistius, translated by Praetextatus; monograph on hypothetical syllogisms by Victorinus. It is very doubtful whether there was a translation or paraphrase of Int.; possibly Apuleius' Peri herm. was used. Instead of Aristotle's Topics, Cicero's was read; Victorinus composed a commentary on it. Cf. on these matters, P. Hadot, Marius Victorinus, Paris 1971.

4 Boethius, In Intr. l a 10; cf. In Int. 2a 93. 5 Boethius, In Intr. 1 a 25: φαντασίας Graeci dicunt, a nobis visa poterunt nominari.

Cicero, Acad. I 40: quam ille φαντασίαν, nos visum appellemus licet. ' In Intr. ed. prima is keyed to Victorinus' version of the Isagoge. 7 The basic source is Boethius, In Int. 2 a 79-80. See further L. M. de Rijk, On the

chronology of Boethius' works on logic, in: Vivarium 2 (1964) 1-49 & 125-162; L. Obertello, Severino Boezio, Genova 1974 ( = Accademia Ligure di Scienze e Let-tere, Collana di monografie I), 1,157 sqq.



288 STEN EBBESEN

(a) a complete set of basic texts, viz. Porphyry's Isagoge, the whole of Aristotle, Cicero's Topics (which, of course, need not be translated), the whole of Plato;

(b) elementary commentaries on each of the basic texts; (c) in some cases, at least, also a more comprehensive commen-

tary; (d) in at least one case, also a paraphrase;8

(e) supplementary monographs, including one demonstrating the compatibility of Aristotelian and Platonic philosophy.

Death prevented him from fulfilling his plan, but he managed to translate the whole Organon (with the possible exception of the Pos-terior Analytics) and to produce several commentaries and mono-graphs. Most of these works are still extant.9

The translations of the basic texts are extremely faithful to the originals which are rendered word by word and morpheme by morpheme10 with a supreme contempt of normal Latin sentence structure.11 The choice of this procedure was very deliberate. Boeth-ius was a consummate master of Latin prose, but he wanted his read-ers to see the real thing.12 The auxiliary works would provide the

8 Boethius, In Int. 2 a 251: huius enim libri post has geminas commentationes quoddam breviarium facimus, ita ut in quibusdam et fere in omnibus Aristotelis ipsius verbis utamur, tantum quod ille brevitate dixit obscure nos aliquibus odditis dilucidiorem seriem adiectione faciamus, ut quasi inter textus brevitatem commentationisque diffusio-nem medius ingrediatur stilus diffuse dicta colligens et angustissime scripta diffundens. atque haec posterius. T o me this looks like a description of a paraphrase, but there have been other suggestions. See D e Rijk, op.cit . 37-38.

' See note *. Surveys in T h e Cambridge History of Later Medieval Philosophy, Cambridge 1982, 53sq., 74sqq., 105. Fundamental studies in L.Minio-Paluel lo, Opuscula, Amsterdam 1972.

10 Thus proba-re = άξιοϋ-ν , proba-mentum = άξίω-μα; see Aristoteles Latinus VI 1-3, Leiden-Bruxelles 1975, 121.

11 Thus the genitive absolute χρησίμης οϋσης της τούτων θ ε ω ρ ί α ς (Porph., Intr. 1,6) becomes an absolute ablative without participle: utili hac istarum rerum specula-tion (Aristoteles Latinus I 6 -7 : 5 ,5 -6) . Also in the commentaries there are occa-sional clumsy renditions of Greek phrases, as In Int. 2 a 13: quae sint negationes cum modo propositionum ... considerator poterit diligenter agnoscere; "cum m o d o proposi-tiones" renders α ί μετά τ ρ ό π ο υ προτάσε ι ς (cf. Ammon., In Int. 8,19; 14,11; 213,33sq.) .

12 Cf. Boethius, In Intr. 2 a 135.




necessary amount of comment and paraphrase. His competence as a translator grew with time, but he never relinquished the principle of strict fidelity to the auctor. For this reason the translations tell us very little about his interpretation of Aristotle.

The commentaries and monographs are more informative, though there is a difficulty in evaluating their testimony because almost everything in them comes from a Greek source. Boethius himself makes no secret of the fact, which can also be established by means of comparisons with extant Greek works. Consequently, a mere recitation of Boethius' words is no sufficient answer to the question, "What did Boethius think about problem X?" . It must also be shown that his use of just those words reflects a choice, i. e. that several sources were available to him or that he sometimes chose to modify his source instead of copying it without change. Thus the fact that he repeatedly stresses that each word has its meaning thanks to a human decision and not by nature13 cannot be taken to show that this is a problem he has given personal thought to unless we can assume that he could have presented another view. If all his sources are lost, we can never establish such a thing. And this is exactly what has been argued in recent times. It has been suggested that the only material at Boethius' disposition was a copy of the Organon with marginal scholia; and that this collection of scholia is no longer extant.14 We may often be able to ascertain the remoter origin of one of the scholia Boethius knew, but we shall never know whether he deviated from his direct source in any way and the stan-dard answer to the question "Why does Boethius say this?" can only be, "Because it was in his only source."

The "one source-no thinking" theory has the support of emi-nent scholars and it cannot be refuted by any means that I can think of. But neither can it be proved by any conceivable means short of finding the supposed manuscript of the Organon with the marginal scholia. T o my mind, the circumstantial evidence in favour of this

13 T h u s In Int. 2 a 23; 54-56; 92-94 ; Intr. Syll. Cat . 763 A; Syll. Cat . 795 A; Divis. 8 8 6 C . Cf . S .Ebbesen, Commentators . . . (see note 1, above), 1,177.

14 J . Shiel, Boethius' Commentaries on Aristotle, in: Mediaeval and Renaissance Stud-ies 4 (1958) 217-244 ; id., Boethius and Eudemus , in: Vivarium 12 (1974) 14-17; id., A Recent Discovery: Boethius' Note s on the Prior Analytics, in: Vivarium 20 (1982) 128-141.



290 STEN EBBESEN

theory, though not negligeable, is less than convincing.15 The observ-able facts are quite as easily explained on the assumption that Boeth-ius had access to several Greek monographs and commentaries and

15 The case for the marginalia and against Boethius' using Porphyry's commentary on the Categories rests on the following assumptions and arguments:

(1) Boethius' Greek copy of the Organon is likely to have looked much like Arethas' from ca. 900 (Vat. Urb. 35), which contains rather ample marginalia. -This is a doubtful assumption. Marginal scholia were known in late antiquity, but papyrological evidence does not suggest that a format like that of Urb. 35 was com-mon then. By contrast, pillaging ancient sources to compose marginal scholia was no unusual occupation for Arethas and his contemporaries.

(2) Boethius himself says that he uses marginalia. For the meaning of In Int. 2 a 250,20-23 "est quidem libri huius . . . obscura orationis series, obscurissimis adiecta sententiis" is that the book has "an obscure course of arguments with highly obscure notes added to i t ."-As far as I can see, the passage means "the doctrines of this book are very obscure, and on top of that the manner of presentation is obscure."

(3) Boethius used one book only for his source. His commentary on the Cate-gories contains a lot of Porphyrien, but also some un-Porphyrian material. Ergo he used one book consisting for the most part of extracts from Porphyry but with some additions from later sources.-I see no compelling reason to grant the first premiss, though I admit that he is unlikely to have consulted a vast literature before writing his commentaries. It may be mentioned that Shiel's guess that all Boethius knew of Themistius' paraphrase of the Topics was a diagram has been disproved by E. Stump, Boethius's Works on the Topics, in: Vivarium 12 (1974) 77-93; cf. S. Ebbesen, Commentators . . . (see note 1), 1,118sq.

(4) Boethius was omnivorous. Yet he does not reproduce all of Porphyry's com-ments on the Categories. Ergo he did not possess a complete copy of that work. - T h e passages which Shiel adduces in support of the first premiss (in: Boethius' Commentaries . . . 233 and 237, see note 14, above) cannot bear the weight of proof. The crucial one is In Intr. 2 a 135, "quocirca multum profecisse uideor, si philo-sophiae libris Latina oratione compositis per integerrimae translationis sinceritatem nihil in Graecorum litteris amplius desideretur." When seen on the background of the preceding lines it becomes clear that in this place Boethius is speaking about a complete and unadulterated translation of Porphyry's Isagoge (and other basic texts), not of scholia.

(5) One medieval Latin manuscript contains Anal. Pr. with scholia, both trans-lated from the Greek by Boethius. This ms. may be taken to mirror Boethius' Greek ms.-There are strong stylistic reasons for attributing the translation of the scholia to Boethius, but also some arguments against doing so. See S. Ebbesen, Analyzing Syllogisms, or Anonymus Aurelianensis I I I - the (presumably) Earliest Extant Latin Commentary on the Prior Analytics, and its Greek Model, in: Univer-site de Copenhague, Cahiers de l'Institut du Moyen-Age grec et latin 37 (1981) 1-20. Yet, even granted that the translator of the scholia was Boethius, it does not follow that the medieval manuscript is a mirror image of one produced by Boethius




that he followed the common practice of using for each work one main source while also exploiting secondary sources. It is an old dis-covery that this hypothesis works well in the case of the commentary on the Categories, the only case in which we still have what may be the main source. Boethius acknowledges a debt to Porphyry16 and actually keeps so close to the latter's extant minor commentary on the Categories (CAG IV 1) that it is simpler to assume that he had direct access to a complete copy of it than to assume second-hand acquaintance by way of a book which also contained the post-Por-phyrian material detectable in Boethius' commentary.

Granted that Boethius' main source was Porphyry's extant work, we can begin to examine the way he used it. As it turns out, he fol-lows his predecessor to the extent of reproducing most of the ques-tions he raised and the answers he gave, but not to the extent of reproducing long segments of his text in direct translation. Boethius expanded arguments which he found too compressed while curtail-ing or suppressing other passages.17 In fact, he followed the proce-dure which his own remarks in this and other works indicate18-and that procedure involved making choices. It looks as if it might be worth while to speculate about his possible motives for choosing as he did.

which again was a mirror image of his Greek manuscript. In fact, there is good rea-son to believe that the scholia in the medieval ms. are just excerpts from a larger collection of scholia, possibly a whole commentary, which had been translated from the Greek and was available to some learned men in the twelfth century. See Ebbesen, op.cit. (this note, above). So, while the scholia may be remnants of mate-rial which Boethius gathered with a view to composing a commentary on Anal. Pr., they cannot tell us how much Greek exegetic material relevant to the text Boethius had access to.

" Boethius, In Cat. 160 A; see note 20, below. 17 Expansion: e.g. Boethius, In Cat.201 D-202B vs. Porph., In Cat. 100,12-16; Boeth-

ius 240C-D vs. Porph. 128,13-15. Curtailment: e.g. Boethius 180C vs. Porph. 86,20-32. Suppression: e.g. Porph.55,3-56,13; 100,23-27. Cf. J.Shiel, Boethius' Commentaries . . . (note 14, above). For the work as a whole, expansion more than neutralizes contraction. PL 64, col. 159-262, which correspond to CAG IV 1, p. 55-142, would require some 125 CAG-pages. Boethius' In Cat. is comparable in bulk to Philoponus' comm. In Cat., and Boeth. In Int. 2 a to Ammonius' In Int.

18 E.g., Boethius, In Cat. 159A; Syll. Cat. 793C; Introd. Syll. Cat.761C. Boethius uses conventional phraseology when saying that he will clarify, expand and abbre-viate. Cf., e.g., Apollonius Dyscolus, De pronomine ( = GG II. 1.1) 3; Themistius, In Anal. Post. 1-2. But this does not imply that he does not mean what he says.



292 S T E N EBBESEN

First of all: why Porphyry? It seems that Boethius turned to Por-phyry not only when composing the commentary on the Categories, but whenever there was a work of his to turn to.19 If Boethius' choice was not determined by unavailability of other books, in what was it grounded? I suggest that it was grounded in the concern for peda-gogy to which he often gives expression20-a concern which he shared with the Phoenician.

It was the general belief of the scholastics that Aristotle himself was consciously pedagogical and that this is apparent from the struc-ture of the Organon21. The books about demonstrative (and, some would add, dialectic) reasoning, i.e. Anal. Post. (& Top.), form the core of the work. But pedagogical considerations motivated Aris-totle to introduce these books with a treatise on syllogistic reasoning as such (Anal. Pr.); and this again with a book on the proximate con-stituent parts of syllogisms, viz. propositions (Int.); and this book again with a treatise on the proximate constituent parts of proposi-tions, i.e. words (Cat.). This, the most elementary book was intended to be read first of all - though the scholastics themselves preferred to warm up with Porphyry's Introduction to the Catego-ries (Isagoge).

When scholasticism was in its infancy, several philosophers had subjected the Categories to unkind scrutiny, pointing out absurd

19 Porphyry was certainly used for the commentaries on Int., almost certainly for Syll. Cat. (and hence for Intr. Cat. Syll.), and Divis.; probably also for Hyp. Syll. The planned work on the concord between Plato and Aristotle presumably was to draw on Porphyry's book on the same subject. For the Topics, Boethius turned to The-mistius. But then there is no evidence that Porphyry ever commented on that work. For the Isagoge, Boethius obviously had to turn to a non-Porphyrian commentary; hence the many un-Porphyrian features of his In Intr. 1 a and 2 a.

20 Thus In Intr.2a 161 & 167; In Cat. 159A, 160A-B (see text in note 53), 250C, 289 C; Syll. Cat. 793C-794C (for better text than Migne's, see De Rijk, op.cit. (n. 7) 30 and note 57 below); Intr. Syll. Cat. 761 C-762C; Divis. 877A. Notice the motives for following Porphyry that Boethius himself gives: In Cat. 160 A: Haec quidem est tempori introductionis et simplicis expositionis apta sententia, quam nos nunc Porphyrium sequentes, quod videbatur expeditior esse planiorque digessimus. In Int. 2 a 7: de interpretatione liber inscriptus est. cuius expositionem nos scilicet maxime a Porphyrie, quamquam etiam a ceteris, trans/erentes latina oratione digessimus. hie enim nobis expositor et intellectus acumine et sententiarum dispositione videtur excellere.

21 Boethius, In Intr. 1 a 12-15; In Cat. 161 B-D; In Int.2a 12 (cf. the scholia in CAG IV 5, p.XL). The Greek sources are numerous. See, e.g., Simplicius, In Cat. 14-15 & 18,14-16; Philoponus, In Anal. Post. 1-2.




doctrines, important omissions, internal inconsistencies and inconsis-tency with other Aristotelian writings. To a large extent this criticism determined the scholastic exegesis. The problems dealt with in the commentaries are very often those raised by the early critics. Many of the supposed deficiencies of the Categories could be explained away by showing that the criticism rested on a wrong interpretation of the text. But not all difficulties could be coped with in that way. Porphyry found a way to interpret the vices as virtues by appealing to Aristotle's concern for pedagogy. Thus the fact that Aristotle does not treat of quality immediately after substance but inserted the chapters on quantity and relation between substance and quality might seem odd both from a Stoic and a Platonist point of view. Besides possible ontological reasons for Aristotle's order, Porphyry mentions that (1) there are more common characteristics between substance and quantity than between substance and quality; (2) in dealing with quantity Aristotle introduced the notions of big and small, which are relative terms; so, in order not to leave the reader in suspense, Aristotle decided to round off with a chapter on relation before going on to quality. Porphyry thus assigns a pedagogical motive to Aristotle, and Boethius accepts the explanation.22 He also follows Porphyry in the conviction that lack of agreement between the doctrine of the Categories and that of other books may be explained as due to pedagogical considerations. Thus they both say that when Aristotle ends his list of types of quality with the remark that there may be more than the ones in the list, this is not a sign of uncertainty, nor can he be blamed for not giving the better list that he has in the Metaphysics. The reason why he does not give it is that he does not wish to perplex the freshmen by introducing intricacies which they need not understand before they have reached a higher level.23 Just in case anybody should be dissatisfied with this explana-tion, Boethius adds another one, this time not from Porphyry, according to which Aristotle said, "And there may be other sorts than the ones I have listed" because he wanted to incite the reader to

22 (1) Porphyry, In Cat. 100,21-23 & Boethius, In Cat. 202C. (2) Porphyry 111,12-15 & Boethius 217 A; Porphyry 127,3-7 & Boethius 239 A. A similar defence of Aris-totle in Porphyry 59,34-60,10 & Boethius 163 B-C.

23 Porphyry, In Cat. 134,25-29; Boethius, In Cat. 252 B. A similar defence of Aristotle in Boethius, In Cat. 289C; cf. Simplicius, In Cat. 427-428; Porphyry not available for comparison.



294 Sten Ebbesen

start t h i n k i n g o n his o w n i n s t e a d o f jus t b e i n g a p a s s i v e r e c i p i e n t o f l earn ing . 2 4 H e r e , t o o , B o e t h i u s d e p i c t s A r i s t o t l e as a c o n s c i o u s p e d -a g o g u e .

P o r p h y r y h e l d t h a t A r i s t o t l e w a s a l s o p e d a g o g i c a l in a p h i l o -s o p h i c a l l y m o r e i n t e r e s t i n g w a y . T h e f i r s t b o o k o f t h e O r g a n o n , h e h e l d , is n o t a b o o k o n w h a t t h e r e is. It is a b o u t an e l e m e n t a r y mat ter , t h e ra t iona l c l a s s i f i c a t i o n o f t h e w o r d s o f t h e v o c a b u l a r y w h i c h re -f l e c t s o u r a w a r e n e s s o f t h e s e n s i b l e w o r l d . T h e c a t e g o r i e s w i t h w h i c h A r i s t o t l e d e a l s s h o u l d b e u n d e r s t o o d as w i d e s t pred ica te s , 'pred icates ' h e r e m e a n i n g w o r d s w h i c h w e m a y a p p l y t o a t h i n g , a n d pr imar i ly w o r d s w h i c h m a y f u n c t i o n as n a m e s o f s e n s i b l e t h i n g s . W e d o b y a b s t r a c t i o n f o r m n o t i o n s o f ( p o s t rem) un iversa le , b u t t o s a y t h a t a c a t e g o r i a l c o m m o n n a m e l ike 'man' s i g n i f i e s a un iversa l M A N jus t m e a n s t h a t t h e r e are s o m e par t i cu lar s e n s i b l e t h i n g s a b o u t w h i c h w e m a y say, " T h i s is a man". 2 5

24 Boethius, In Cat. 252 C: est quoque alia causa ut nos ad exquirendas alias qualitates non solum propriorum suorum doctor [doctorum Migne] sed etiam nostrorum aliquid inveniendi incitator admitteret. Cf. Ammonius, In Cat. 88,20-22; Philop., In Cat. 156,8-11. Ammon. and Philop. do not mention Porphyry's explanation. Simplicius, In Cat. 264,2-4 has it, but does not mention the explanation of Ammon. & Philop. Cf. Boethius 294 Α-B where he uses a similar explanation, also found in Philop., In Cat.205,26-28. Porphyry and Ammon. are not available for comparison in this place.

" See, in particular, Porph., In Cat. 56-58 & 90-91. Read 56,8-13 as follows: πασα άπλή λέξις σημαντική, δταν καθ ' ού σημαίνηται πράγματος άγορευθη τε και λεχθ-fj, λέγεται κατηγορία, οίον δντος πράγματος τοΰδε τοΰ δεικνυμένου λίθου, ού άπτόμεθα ή δ βλέπομεν, δταν εΐπωμεν έπ' αύτοΟ δτι λίθος έστίν, ή 'λίθος' λέξις (κατηγορία, τό δε πραγμα) κατηγόρημά έστι. σημαίνει γαρ τό τοιόνδε πραγμα και άγορεύεται κατά τοΰ δεικνυμένου πράγματος λίθου. Cf. Simpl., In Cat. 11,2-3 & 17,5-7. Boethius does not reproduce the two crucial passages on the meaning of κατηγορία, κατηγόρημα, and κατηγορεΐσθαι (Porph., In Cat. 56,6-13 & 58,16-18). Yet cf. Boeth., In Cat. 162 D: Quoniam rerum prima decern genera sunt, necesse fuit decern quoque esse simplices voces quae de subiectis rebus dicerentur: omne enim quod significat, de ilia re dicitur, quam significat. Syll. Hyp. p. 210 Obertello: in praedicativa [he is speaking about 'homo animal est'] igitur id spectabimus quod ipse homo animal sit, id est nomen in se suscipiat animalis... Itaque praedicativa propositio rem quam subicit praedicatae rei suscipere nomen declarat. Introd. Syll. Cat. 768 C: subiectum est quod praedicati suscipit dictionem. See also Boeth., In Cat. 243 C where he is eager to make it clear that the fact that genus is entailed by species means that ubicumque species sit, mox quoque nomen generis praesto est; ubi autem sit genus non necessario speciei vocabulum sequitur. The words immediately before strongly resem-ble Iamblichus, apud Simpl., In Cat. 230,32-231,1; but ubicumque-sequitur is not paralleled there. Porphyry is not available for comparison.




Porphyry had predecessors in the endeavour to make the Cate-gories acceptable to Platonists by restricting its scope to the sensible world. But he was particularly insistent on the point, and Boethius followed him in this.

Critics of Aristotle had observed that Aristotle stood the world on its head by calling the sensible particulars p r imary substances. Porphyry would not deny that if Aristotle were doing basic meta-physics, this would indeed be standing the world on its head. But, he held, the point is that in the Categories Aristotle is not working on such a deep level. On the level on which he is working, it is perfectly reasonable to call sensible particulars p r imary substances while call-ing secondary the abstract universals whose only being consists in being predicable of the first ones, and to leave such intelligible enti-ties as God or Mind totally out of account. Boethius agrees.26 That he did not only copy but also understand this Porphyrian view may be inferred from the fact that he manages to say the same things as his source without reproducing it verbatim. He even manages to improve upon the master's formulations. Thus it is Boethius who says that "secondary substances do not have their being except in being predicable" and that "what causes secondary substances to be is nothing but their predicability of primary substances." Porphyry uses a less pregnant formulation, saying that the particular animals and men are the causes of being for the commonly predicated ones.27

So, Aristotle was right in abstaining from deeper matters when introducing people to philosophy. Porphyry extended the principle of abstention from deeper matters and concern for the novices from being a principle of interpreting Aristotle's logic into a rule guiding his own activity as a commentator-at least when he was writing for beginners, as was the case when he composed his Introduction to the Categories and the extant commentary on the same work. On the whole, he remains faithful to the programme announced in the beginning of the Introduction: to give a concise account of Peripa-tetic lore from a logical point of view, abstaining from the deeper

» Porph., In Cat. 90-91. Boethius, In Cat. 183-184. 27 Boethius, In Cat. 185 D: secundae substantiae sunt quae in subiecto non sunt et de sub-

iecto praedicantur. Ergo esse suum nisi in hoc quod de aliquo praedicantur non retinent. praedicantur autem secundae substantiae de primis; ergo, ut secundae substantiae sint, praedicatio de primis substantiis causa est. Porph., In Cat. 90,33-91,1: τά καθ ' εκασ-τον ζ φ α . . . α δή και α ί τ ια τοις κοινή κατηγορουμένοις έστι τοϋ είναι .



296 STEN EBBESEN

issues, such as the mode of being of universals, keeping in mind that he is writing an introduction.28 Though Boethius, when commenting on this passage, could not quite keep himself from telling what Por-phyry abstained from saying,29 he elsewhere makes the Porphyrian programme his own.

As a matter of fact, Porphyry tried to make Aristotelian logic function on a minimum of assumptions, notably these:30 (1) that man can recognize certain individuals in the sensible world; (2) that man can name these individuals; (3) that man can recognize a similarity between several individuals; (4) that man can decide to use a certain sound as a name of any individual which is similar to others in respect of a certain feature; (5) that man can recognize a similarity between certain names; (6) that man can decide to use a certain sound to name any individual name which is similar to others in respect of a certain feature.

These are the basic assumptions needed for an interpretation of the Categories. They allow for the creation of our primary object-language with its universal names, the classification of these names in ten categories, and for the creation of a second-order language, which is the subject of Peri hermeneias and which is necessary for talking about the first-order language which is the subject of the Categories.

Some more assumptions are needed to deal with propositions. These are basic: (7) that any two men confronted with the same sense-data will isolate the same individuals and recognize the same similarities; (8) that the thoughts or "concepts" to which these "recognitions" give rise can be stored and recalled to consciousness when required; (9) that simple concepts can be combined into com-plex ones; (10) that man can invent words indicating the internal syntax of complex concepts; (11) that one man can teach another to use words in the same way as himself; (12) that by uttering the words associated with the simple or complex concept presently actu-alized in him one man can actualize the same concept in a fellow-man who speaks the same language. These further assumptions

28 Porph., Intr. 1. 29 Boethius, In Intr. 1 a 24 sqq.; In Intr. 2a 159 sqq. 50 Cf. S.Ebbesen, Commentators . . . (see note 1, above), 1,141 sqq. It is pertinent to

notice that Boeth., In Int. 2 a is one of the really important sources for Porphyry's thought.




allow men to communicate with each other by means of propositions analyzable into subject, predicate and their mutual relations, without the need of pointing at the object of discourse.

Of course, this is not the whole story, but Porphyry did his best to explain Aristotelian logic without a host of other assumptions, and he tried to be a good pedagogue by not introducing an assump-tion before it was needed. Thus, in spite of the crucial role of con-cepts in his Aristotelian semantics, he did not formally introduce concepts in the little commentary on the Categories. Concepts could wait until the reader got to the Peri hermeneias. Boethius followed him in this. But he did more than just following Porphyry.

Boethius demonstrates his understanding of Porphyry's thought by moving the presentation of the fundamental theory about the ori-gin of our language to the very beginning of his commentary on the Categories. Porphyry operated with two labellings of things, two "impositions of names".31 By the first imposition man has created his object-language, by the second his meta-language. Using and pre-senting the model requires some delicacy. For instance, it may be wise to de-personalize the first impositor as much as possible so as not to end up with one of the following ridiculous scenarios (with which some people have ended up):

(a) One wise impositor at some time decided what things ought to be called.

(b) A congress of terminologists convened to invent language. Having done so, they went home to meet again at a later time in order to perform the second imposition.32

Boethius avoids these absurdities. Porphyry had de-personalized the impositor by calling him 'man himself', not 'some man'.33 Boeth-ius achieves the same by saying that only the race of men could put names on things and that the human mind applied words to the things.34 On the other hand, it is equally absurd to let an abstract mind perform the naming. Some particular man must have been the first to say 'man'. So it is reasonable enough when Boethius-not

31 Porph., In Cat. 57-58. 32 See S. Ebbesen, op.cit., 1,178. 33 Porph., In Cat.57,21. 34 Boeth., In Cat. 159A. Cf. In Int .2a 55,1-7.



298 S T E N E B B E S E N

quoting Porphyry-speaks of the man who first said 'man'.35 It is less reasonable that he sometimes (not in the commentary on the Cate-gories) suggests that the imposition must have resembled a christen-ing ceremony, the impositor saying, "Let this be called a man".36

Apart f rom the fact that experience shows that there are other ways of introducing new words, the picture Boethius paints of the imposi-tion ceremony is based on the unnecessarily strong and unconvin-cing claim that all names start as proper names. The model of the first imposition is designed to show that our use of generic names is dependent on a confrontation with the sensible particularized world. T o avoid solipsism, the imposition must in some way be a public event, but there is no need to deny the possibility of a telescoped procedure, so that the impositor can say, "Let such a thing be called a horse". Boethius himself may on second thoughts have found the christening ceremony unconvincing. In the first version of his opus-cule on categorical syllogisms, the Aristotelian claim that the nomi-native is more truly a name than the oblique cases of nouns is defended with the remark that the man who first introduced the word circus probably said, "Let this be called circus"?7 In the revised version, in which he does much to eradicate crude expression and thought, this explanation has been dropped. Instead we find the rather more sophisticated remark that since we start by sensing things which are present to us, and it is clear that men have assigned their words to the things which we perceive with a sensation in the pres-ent, it is reasonable to say that properly speaking verbs always have present signification, so that Aristotle's characterization of past and future verb forms as non-verbs is justified.38

On the whole, then, Boethius' account of the theory of imposi-tion is good. Moreover, he puts the notion of second imposition

35 Boeth., In Cat . l83D. 34 Boeth., In Int. 1 a 46; In Int. 2 a 64; cf. next note. B. may have had Greek models for

the christening ceremony. Cf. Anonymous Commentary on Aristotle's De Interpre-tatione (Codex Parisinus Graecus 2064), ed. L.Tarän, Meisenheim 1978, 3 ( = Beiträge zur klass. Philol. 95).

37 Boeth., Syll. Cat. 796 A; cf. preceding note. 38 Boeth., Intr. Syll. Cat .765D-766A. Notice that in Intr. Syll. Cat. Boethius is less

dependent on his Greek source(s) than in Syll. Cat. Thus he expands the sections on the parts of speech, 796 C - D , so as to present the whole Latin inventory of eight parts, one being interjections; see 766 B-C.




words, or names of names, to use in a place in which his main source did not. Discussing the transitivity of predication he has to deal with the example 'Man is a species, Socrates is a man, ergo Socrates is a species'. Boethius solves the paralogism by pointing out that 'species' does not occur in the definition of man, and so 'man is a species' cannot be a predication 'in quid'. No, he says, the fact is that in this proposition the function of 'species' is to indicate that 'man' is predi-c a t e of individuals only "and ['species'] is, in a sense, a name of names".39 I.e., the proposition may be paraphrased as "the word 'man' is predicable of individuals only", and 'species' is the name of those names which are predicable of individuals only. Boethius owes this solution of the paralogism to a Greek thinker,40 but not to Por-phyry's minor commentary on the Categories. His departure from the usual source probably means that he found the explanation attractive-as indeed it is in the theoretical framework within which he is working.

The cautious formulation that genus and species are "in a sense" {quodatnmodo) names of names may indicate his own and/or his source's reluctance to call 'species' a name of names because that ex-pression was primarily used of the designations of the parts of speech-noun and verb.41 But there may also have been a more seri-ous reason for his cautiousness.

At this point, after so many nice words about Boethius' intelli-gence, I wish to make a digression about an intricate matter which he did not manage to sort out and which shows a limitation of his ability to use in one place what he had learned in another context.

When on his own, Boethius was not very well at ease when hav-ing to explain the sort of relations that obtain between such non-categorial terms as 'species' and categorial terms like 'man'. This is very clear in his two works on topical argumentation (In Topica Ciceronis, De differentiis topicis). For these he had no good Greek sources but had to produce his own mixture of a Greek theory which he knew from Themistius-an eloquent man, but not the greatest of philosophers-and of Roman rhetorical tradition. The result is astonishing in several respects. One cause of wonder is the

39 Boeth., In Cat. 176D-177A. 40 Possibly Iamblichus. See Dexippus, In Cat. 26; cf. S.Ebbesen, op.cit. 1,231. 41 See Boeth., In Cat. 159B-C, corresponding to Porph., In Cat. 57,29-58,3.



300 STEN EBBESEN

total absence from these works of the notion of names of names, although it would appear to be highly relevant in the context, as some of Boethius' medieval commentators saw.42 For in the topical works he operates with second-level words or concepts (it is not quite clear which), such as genus, species, definition. Associated with each of these is a set of axioms, such as 'That to which the definition of the genus does not apply is not a species of that of which the definition in question is a definition.' This particular axiom is sup-posed to bail for the soundness of the inference, 'an animal is an ani-mated sensible substance, a tree is not an animated sensible sub-stance, ergo a tree is no animal'. Directly the argument is supposed to hinge on the axiom, indirectly on the notion of definition. In other words, to decide the question whether trees are animals we may construct an argument founded on the notion of definition and on a particular instance of a definition, the definition of animal.43

But Boethius has a worry.44 Are not definition and definiendum really the same thing? Did not the definiendum, animal, occur as a term in the proposition we wanted to prove? Isn't then our use of the definition of animal a case of using a term to prove something about itself? To explain that nothing is wrong Boethius uses singularly obscure language, a vice to which he was not addicted. I think what he wants to say is that the definition 'animate sensible substance' may be considered in three ways: (1) as the thing or the sign of a thing which is really identical with animal though notionally differ-ent; (2) as a phrase; in this sense the definition is also a thing "in so far as phrases are things"; (3) as a definition; it is in this last sense that it is a locus (τόπος), i. e. a source of arguments (sedes argumento-rum, αφορμή έπιχειρημάτων), and in this sense it is not a thing. This is suggestive, but it certainly does not attest to a clarified con-ception of the semantics of non-categorial terms.

Boethius knew next to nothing about Stoic logic. Porphyry in his Aristotle commentaries often compared Stoic terminology and tenets with Aristotle's. Boethius rarely bothered to include such passages in his own commentaries. He found them both irrelevant and 42 See N.J. Green-Pedersen, The Tradition of the Topics in the Middle Ages, Mün-

chen 1984. 43 Boeth., Diff. Top.II, 1187A-B. 44 Boeth., In Top. Cie.III, 1083Csqq.; 1091 Β; cf. I,1055sqq. I owe my awareness of

the problem of these passages to N.J. Green-Pedersen, op.cit.




obscure.45 Real knowledge of Stoic logic had disappeared long before his times, but fragments-often distorted fragments-of Stoic theory would occur in all sorts of technical literature, not least in rhetorical writings. I think Boethius' confusion about the status of definitions was occasioned or increased by his use of a source that contained some distorted version of a Stoic distinction between cor-poreal and incorporeal entities, such as words vs. their meanings; or between incorporeal somethings (τινά) and nothings (οϋτινα), such as predicates vs. universals.

Boethius' works on topics also contain a distinction between argumentatio, which is a proof qua (pronunciation of a) string of words, and argumentum, which is the sense of the argumentatio.46

This must reflect a Stoic distinction between an incorporeal έπιχείρημα and a corporeal έπιχείρησις, comparable to many other Stoic distinctions between corporeal entities with names in -ια or -σις and their incorporeal effects or meanings which have names in -μα.47 The purpose of the distinction was no longer clear to Boeth-ius.48 Just very tentatively would I suggest that there may be a con-

45 See Boeth., In Int. 2 a 24; 71; 201. 46 Boeth., In Top. Cic.I,1050B; 1053A-1054A; Diff. Top. 1174D; 1183A. " Cf., e.g. κατόρθωσις-κατόρθωμα, v. Arnim, SVF III, nos.85 & 524. ωφέλεια

-ωφέλημα, αμαρτ ία-αμάρτημα are other examples. This sort of distinction also underlies Porphyry's between κατηγορία and κατηγόρημα (see note 25, above).

48 This appears both from the fact that he makes little use of the distinction and from the discussion in In Top. Cie. 1,1053 A-1054 A. The first evidence of the distinction argumentum-argumentatio is Cicero, Part. Or. 45: argumentationem quaerere videris, quae est argumenti explicatio·, cf. Cie., Inv. I 40,74. See also Fortunatianus, Ars Rh. II, §23 & 28 ( = Halm, Rh. Lat. Min. 115; 118); Victorinus, In Rh. Cie., I 29; 31; 40 ( = Halm, op.cit. 231; 232; 240; 247); Cassiodorus, Inst.II, p. 105 Mynors. I have not found the distinction between έπιχείρημα and έπιχείρησις in any Greek source. But notice the quotation of Porphyry in Syrianus' commentaries on Her-mogenes, Rabe, Rh. Gr. XVI 1, p. 93 & XVI 2, p. 14: toö μεν γαρ λόγου ψυχήν δοκοΟντος εχειν και σώμα, ή μεν τών νοημάτων ευρεσις δικαίως άν ψυχή τοΟ λόγου νομίζοιτο, ή δε έρμηνεία σώμα. Compare Boeth., Diff. Top. I, 1174D: Non vero idem est argumentum et argumentatio: nam vis sententiae ratioque ea quae clauditur oratione, cum aliquid probatur ambiguum, argumentum vocatur; ipsa vero argumenti elocutio [ = έρμηνεία] argumentatio dicitur. Quo fit, ut argumentum quidem virtus et mens [ = νοΟς or ψυχή ?] argumentationis sit atque sententia, argumentatio vero argumenti per orationem explicatio. Similarly In Top. Cie. 1,1053 B: Aut enim elocutio et contextio ipsapropositionum... argumentatio vocatur, argumentum vero mens et sententia syllogismi. Syrianus also has a definition of έπιχείρημα closely related



302 STEN EBBESEN

nection between the two detached fragments of Stoic theory. The original idea may have been that just as physical τόπος is an incor-poreal something, so is a τόπος in argumentation theory. Its con-tents are other incorporeal somethings, viz. argumenta (επιχειρή-ματα), which constitute the meaning of argumentations (έπιχειρή-σεις), our verbal formulations of proof.

Boethius understood Porphyry much better than he understood the Stoics. He understood Porphyry's de-ontologizing of logic and his economy of assumptions so well that on occasion he refused to follow his teacher when the master forgot his own principles.

Porphyry suggests that the order in which Aristotle deals with the principal categories, viz. substance-quantity-relation-quality, may be due to the fact that (1) being a body is prior to being a quali-fied body, and being a body implies having dimensions, and so implies having quantity; (2) when the dimensions pre-subsist, bigger and smaller, which are terms of a relation, accrue to them. Boethius reproduces this explanation. But he performs a tiny little change. The statement that the relation of bigger and smaller accrues to the pre-subsisting dimensions becomes, "quantity being posited we also have the relation of bigger and smaller."49 The removal of the ontologically loaded word 'pre-subsisting' may be significant.

Porphyry sinned in an even graver way against his own principles when he suggested that Aristotle was wrong in claiming that the known may exist prior to the knowledge of it. Perhaps, Porphyry said, it is best not to look at knowledge as something residing in men, but as residing in the nature of the things that may be known, since the eternal mind understands all things that are, whereas the knowledge of the separate things always descends to men. So that when there is something sensible, there is a universal sensation; and

to the Ciceronian/Boethian. Compare Boeth., Diff. Top. 1174D Argumentum est ratio rei dubiae faciens fidem with Syrianus, op.cit. X V I 1, p. 57 επιχείρημα δε έστι λόγος προς πίστιν τοϋ υποκειμένου ζητήματος παραλαμβανόμενος.

4 ' Porph., In Cat. 100; 111; 127, matched by Boeth., In C a t . 2 0 2 B - C ; 2 1 6 D ; 239A. Compare Porph., 111 ,9-10 προυφεστηκότος μήκους πλάτους βάθους έπιγίνεται τό μείζον και τό ελαττον, απερ έστιν πρός τι with Boethius, 216 D posita quanti-tate maius [magis Migne] minusve esse necesse est. Quare, cum quantitatem continuo ad aliquid consequatur, recte post quantitatem relativorum series ordinata est. See also the appendix to this paper.




when there is something that may be known, there is a universal knowledge, the terms of the relation thus being simultaneous.50

Porphyry here uses the Neo-Platonist device of grounding the possibility of communication between distinct entities in a higher-level union of these distinct entities. And he introduces the eternal mind so conspicuously absent from the rest of his exegesis of Aris-totle's logic. Boethius deserts his main source at this point. Only at the end of a longish defence of Aristotle's opinion does he mention that some, including Porphyry, disagree, and then goes on to quote Porphyry, excusing himself for doing so with the remark that this will not take up much space.51 But the work he quotes from is not his usual source, Porphyry's minor commentary on the Categories. It must be the big one Ad Gedalium. The quotation is likely to be sec-ond-hand, but this is immaterial. The question is: what did Boethius gain by switching source? Apparently very little, for the quotation says much the same as the minor commentary and there is no ap-preciable gain in terms of brevity. Yet he gains one thing: he avoids introducing the Eternal Mind in a work of logic, for the extract from the big commentary contains no mention of the Eternal Mind.

But is it really so remarkable that Boethius is Porphyrian, and sometimes a bit more than Porphyry himself? Yes, it is. For not every-body was so. In particular Iamblichus was not. He had introduced the two-level commentary, consisting for one part of the pedestrian Porphyrian exposition, for another of an "intellectual" interpretation of the same matters. Iamblichus had the peculiar idea that Aristotle had plagiarized the Pythagorean Archytas when writing the Catego-ries, and that the true interpretation of the doctrine of categories had to be Pythagorean. Boethius knew of the existence of Iambli-chus' commentary.52 We cannot be certain that he had read it. But if he had not, he had read some other commentary depending on Iam-blichus, for he knew the sort of philosophizing about the categories which it contained. And he liked it.

In his own commentary Boethius, after presenting Porphyry's explanation of what the Categories is about, adds that he has chosen to follow Porphyry because he is so easy to understand for a begin-

50 Porph., In Cat. 120,33-121,3. 51 Boeth., In Cat.233B-D. " See Boeth., In Cat. 162 A; 224 D & 225 B.



304 S t e n Ebbesen

ner in philosophy. But, says Boethius, I have in mind to write another work in which I shall treat of three questions, one of which is what the Categories is about. The explanation that I shall give there will be very different from the one I give here, much pro-founder and in accordance with the advanced theory of the Pytha-goreans.53

It is not obvious what the other two questions are that Boethius intended to write of elsewhere. Various solutions have been prof-fered.54 I would like to point out that there are two more references in Boethius' commentary to a future work. In one place he promises to discuss the authenticity of the work on categories that Iamblichus believed to be by the old Pythagorean Archytas, the contemporary of Plato.55 In another place he promises a refutation of people who criticize Aristotle's list of categories for being too long, too short, or badly ordered.56 These may be the two missing questions. At any rate, there is every reason to believe that Boethius wanted to treat them in the same work as the first one.

53 Boeth., In Cat. 160A-B. Migne's text is corrupt . De Rijk, op. cit. (note 7, above) 133 quotes a Brussels ms. I here print the text of ms. T h o t t 168,2°, ff . 2 v -3 in the Royal Library of Copenhagen. In the essentials, there is agreement with the Brussels ms. Est vero in mente de tribus olim quaestionibus disputare, quarum una est quid Praedica-mentorum velit intentio, ibique numeratis diversorum sententiis docebimus cui nostrum quoque accedat arbitrium. Quod nemo huic inpraesentiarum sententiae repugnare mire-tur, cum videat quanto ilia sit altior, cuius non nimium ingredientium mentes capaces esse potuissent, ad quos mediocriter inbuendos ista conscripsimus. Efficiendi [sic!] ergo et quodam modo disponendi mediocri expositione sunt in ipsis quasi disciplinae huius

foribus quos adhanc sententiam paramus admittere. Hanc igitur causam mutatae senten-tiae utriusque operis lector agnoscat, quod illic ad scientiam Pythagoricam perfectamque doctrinam, hie ad simplices introducendorum motus expositionis sit accommodata sen-tentia.

54 De Rijk, op. cit. 137 suggests σκοπός , χρε ία , έ π ι γ ρ α φ ή . A scholiast in ms. T h o t t 168,2°, f . 2 v says: Hoc non ita est intelligendum quasi alteram fecerit expositionem, quia quod hie promittit in secundae editionis libri Perierminias prima parte evidenter exponit, ubi et in fine ita concludit: "Et de intentione quidem et de libri inscriptione et de eo quod hie maxime Aristotelis liber esse putandus est, haec dicta sufficiunt." T h e words in quotes come f rom Boeth., In Int. 2 a 13,9-11. Above 'diversorum senten-tiis' the scholiast writes: Aspasii, Alexandri, Theophrasti, Porphyrii, Aristotelis, Andronici. T h e scholiast is hardly right in thinking that Boethius is referr ing to In Int. 2 a 7-13, but it is very probable that Boethius there used some of the material he had intended fo r use in the promised work .

55 Boeth., In Cat. 162 A. " Boeth., In C a t . l 8 0 C .




These passages show that Boethius had a clear notion of levels of interpretation and that his exclusion of Neo-Platonic speculation from the elementary commentaries was deliberate. But isn't he cheat-ing his readers by offering them an interpretation which he himself considers only second-best? Hasn't he a better defence for his proce-dure than the claim that the profound theory is too difficult for novices? I think he has, though he does not directly confront the question. He does however, in his Introduction to the Categorical Syllogisms, tell his readers not to think that in teaching them logic and its way of dealing with language he wants them to unlearn the grammar they were taught at school. What the reader is invited to do is to consider the same matter from a new angle which gives a profounder understanding. Looking at the constituents of speech from a grammatical and a logical point of view may be compared to looking at a line or a surface from a mathematical and a physical point of view. The several disciplines which study the same object stand in no conflict. On the contrary, he says, a true understanding of nature can only arise as a result of using more than one approach.57

A similar attitude to the relation between logic and profounder theories would hold that Aristotelian logic as interpreted by Por-phyry is a perfectly legitimate and useful discipline. We must have a theory of the linguistic behaviour associated with our awareness of the sensible world, and it is an advantage if this theory carries few ontological assumptions. The profounder theory adds a new dimen-sion to our understanding of the world. It does not render the shal-low theory superfluous.

57 Boeth., Introd. Cat. Syll. 7 6 1 D - 7 6 2 C . Notice that 761 D Idem namque-762C explicata cognitio is a replacement of the original exhortation to look up Boethius' Greek models; Syll. Cat. 794C (I quote f rom ms. Tho t t 166,2°, f. lv) : Et hi quidem, si nos, ut arbitror, non sufficimus, eos commentaries de quibus haec nos protulimus degustent: blando forte sapore subtilitatis eliciti, quamvis infrenes et indomiti certatores sint, veterum virorum inexpugnabilibus auctoritatibus adquiescent. Siquis vero Graecae orationis expers est, in his, vel si qua aliorum sunt similia, desudabit. Itaque haec huius prooemii lex erit, ut forum nostrum nemo non intellecturus, et ob id culpaturus inspici[ci]at. Se ne prooemiis nihil afferentibus tempus teratur, inchoandum nobis est, illo prius depulso periculo ne a quoquam sterilis culpetur oratio: non enim eloquentiae compositiones sed planitiem consectamur; quare, si hoc efficimus, quamlibet incompte loquentes, intentio quoque nobis nostra perfecta est.



306 STEN EBBESEN

What was the profound work to be like? I think we have enough information to gaiii an impression of what a true Pythagorean inter-pretation of the categories would be for Boethius. The story would run more or less like this:58

58 That, broadly speaking, 'Pythagorean' must mean "after the manner of Iamblichus" is obvious, as pointed out by De Rijk, op. cit. 136 sqq. A more precise idea of what Boethius would understand by 'Pythagorean' in connection with the Categories may be obtained from the following observations, (a) Simplicius and Syrianus con-nect the list of exactly ten categories with Pythagorean speculation (Archytas). See Simpl., In Cat. 13,21-23; 51,3-4; 68,22-28 ( = Ps.-Archytas, ed. Szlezäk, 57); Syr-ianus, In Hermogenis Π. στάσεων (Rabe, Rh. Gr. XVI 2) 58. So does Nicomachus, Intr. Arithm. II 22 (pp. 122-123 Hoche): . . . τον δέκατον αριθμόν κατά τό τοις Πυθαγορικοΐς δοκοΟν ώς τελειότατον, καθ' δν και αί δέκα σχέσεις ώφθησαν ήμϊν προ βραχέος ποσότητα λαμβάνουσαι και αί δέκα λεγόμεναι κατηγορίαι και τών ημετέρων χειρών καϊ ποδών αί τών ακρωτηρίων διαιρέσεις και σχέσεις και ετερα μύρια. Boethius, Inst. Arithm. II 41 (p. 139 Friedl.) expands this as fol-lows, . . . propter denarii numeri perfectionem, quod erat Pythagorae conplacitus ... Inde etiam in Aristotelica atque Archytae prius decern praedicamentorum descriptione Pytha-goricum denarium manifestum est inveniri; quando quidem et Plato, studiosissimus Pythagorae, secundum eandem disputationem [presumably = κατά τον αύτόν λόγον, thus betraying a Greek source for this addition to Nicomachus] dividit, et Archytas Pythagoricus ante Aristotelem, licet quibusdam sit ambiguum, decern haecpraedicamenta constituit. Inde etiam decern membrorum particulae, inde alia permulta, quae omnia per-sequi non est necesse. (b) Boethius' contrasting Porphyry's interpretation with the pro-founder Pythagorean one presupposes that the latter treats the categories as real, intelligible entities. And so it did according to Simpl., In Cat.68,22-28; 91,14sqq. (c) Simplicius and Dexippus consider it a hall-mark of Pythagorean exegesis of the Categories to hold that there is a natural link between things and names. See Simpl., In Cat. 13,21-26; 40,5-13; 105,3-4; Dexippus, In Cat. 16,33-17,3 (notice the kick at Iamblichus in 17,4-6). (d) Proclus' exposition of Pythagorean semantics in his scholia on Cratylus XVI (ed. Pasquali pp. 5-6) is very closely related to the (Iambli-chean) account of semantics in Simpl., In Cat. 12,12-13,11. That semantics does postulate the natural link between names and things, and introduces the notion of original unity in the (Divine) Mind. It also links up perfectly with the distinction between two senses of φύσει presented in Proclus, op. cit. XVII, pp. 7-8 Pasquali and Ammon., In Int. 34 sqq. (e) Boethius, In Int. 2 a 21-23, actually sketches this sort of semantics, though without attributing it to any particular school. A central passage, 22,6-13: si quis ad naturam redeat eamque consideret diligenter, agnoscet cum res est, eius quoque esse intellectum: quod si non apud homines, certe apud eum, qui pro-priae divinitate substantiae in propria natura ipsius ret nihil ignorat. et si est intellectus, et vox est; quod si vox fiterit, eius quoque sunt litterae, quae si ignorantur, nihil ad ipsam vocis naturam. Cf. L. M. de Rijk, Boece logicien et philosopher ses positions semantiques et sa metaphysique de l'etre, in L. Obertello (ed.), Congresso internazio-nale di studi Boeziani-Atti, Roma 1981, 141-156. (f) From Nicomachus' Arithm. I 1-6, which Boethius himself had rendered in Latin, he would know to connect the




The divine mind, which is the wisdom of which philosophy is love, may also be called the primordial number. It contains the multi-tude of intellectual ideas but is characterized by non-distinction between mind, mental grasping and the mentally grasped. It gives the measures of being to all things. On this exalted level there is no need of names. But on the lower level of soul it is different. The soul is not the things themselves. It has "pictures" of the things. These pic-tures are concepts or names. They exhibit the things as structured by the primordial number which (possibly by means of oppositions like even/odd) has structured the things in the perfect number of ten categories-ten like the fingers of the hands and the toes of the feet. The bottom is reached when the soul's immaterial names are embod-ied in matter, in sound. Only wise men who see the Mind and the nature of things can choose the appropriate matter for the name-forms or meanings, and thus perform this final act in the process that creates human language. But even wise men do not always see the same aspects of the things to be named. They are like artists painting the same motif. Even good artists produce different pic-tures. Consequently it is no wonder that different peoples may have different words for the same, or that even the same people may have more than one name of the same thing. The various names are the expression of the givers of names attending to different aspects of the forms of things. Language is "by position" (θέσει) in the sense that the phonic stuff constituting the matter of each name was chosen by somebody. By nature (φύσει) in the sense that the matter was not chosen at random but in imitation of the form to be con-veyed by it. And thus there is a path leading back from material names to concepts and thence to things themselves and the original

theory of divine exemplar ideas with the name of Pythagoras, and also the 'Pytha-gorean' explanation of what philosophy is. This explanation occurs with ascription to Pythagoras in Ammon., In Intr.9,7-23, in a list of definitions of philosophy. The corresponding passage in Boeth., In Intr. 1 a 7, contains no list, but just the 'Pytha-gorean' explanation, without ascription, (g) Addressing Philosophy, Boethius in the Consolation 1 pr. 4,38-39 says, Instillabas enim auribus cogitationibusque cotidie meis Pythagoricum illud επου θεφ. Nec conveniebat vilissimorum me spirituum praesidia captare, quem tu in hanc excellentiam componebas ut consimilem deo faceres. (h) The passage on levels of cognition in Cons. 5 pr. 4,25 sqq. is consistent both with the 'Pythagorean' semantics and with Boethius' program of not penetrating to the pro-foundest level when writing about logic for beginners.



3 0 8 STEN EBBESEN

unity. Following that path the philosopher complies with Pythago-ras' admonition επου θεφ.

Darkness fell over the world of learning soon after Boethius' death. But his writings survived and were influential in the early phases of the medieval renaissance of philosophical studies. By keep-ing, for the most part, a clear distinction between the pedestrian and the profound level of interpretation, and by only handing the first down to posterity, Boethius greatly contributed to the sanity of medieval logic. Listen to this passage from Iamblichus:59

"Since the power of the one from which [the one] everything that has quantity is engendered extends self-identical all through things and limits each thing by proceeding from itself, insofar as it extends totally indivisibly all through things, it establishes the con-tinuous, and insofar as it exercises its procession as one, indivisible procession without delimitation. But insofar as it makes a stop in its procession at each of the forms and insofar as it limits each of them and makes each of them one, in so far does it produce the discrete. In virtue of the one and most principal cause which simultaneously comprises these two acts, it produces the two quanta. And in virtue of its total identity everywhere, both in each of the parts and in them all, it effects the continuous, but in virtue of the identity of each of them to themselves and of its being in its totality in each one it engenders the discrete. And in virtue of the mutual union of the intelligible quanta it establishes the continuous; but in virtue of their mutually separated union it establishes the discrete. And by virtue of its static act it creates the discrete, by virtue of its proceeding act it creates the continuous. Since, then, it is both static and proceeds, it engenders both. For the power of the intelligible measures simulta-neously contains both those quanta which are static and those which proceed, in one and the same."

By not giving the early medieval logicians a Pythagorean com-mentary on the Categories Boethius saved them from being led astray by such gibberish.

5 9 Apud Simpl., In Cat . 1 3 5 , 1 0 - 2 6 .




Appendix A new fragment of Porphyry's Ad Gedalium?

In his commentary on the Categories Boethius accepts Porphyry's defence of the order substance-quantity-relation-quality. But in De Trinitate, ch.4 1252 A he uses the order advocated by Ps.-Archy-tas and his followers, viz. substance-quality-quantity-relation. But then, in De Trinitate he is working on a profounder level.

The discussion about the order is not an idle exercise in the art of squeezing juice of a stone. It is about the fundamental questions of ontology, and accordingly Aristotle's commentators paid great attention to it. In the minor comm. Cat., Porphyry argues that Aris-totle's order is not only pedagogically, but also ontologically/epis-temologically sound.60 His central claim is that for something to be at all requires quantity but not quality. This claim makes good sense in the context of Porphyry's general interpretation of the Categories as dealing with words predicable of sensible things. On this interpre-tation, Aristotle does not want to dig down to deeper ontological levels. The subject, then, in which he says that the accidents are is not prime matter or intelligible substance, but individual sensible substance composed of matter and form.61 This notion of substance entails the notion of quantity, but not of quality in the sense in which Aristotle speaks of qualities in the Categories; for they are accidents, not substantial forms.62

In one of the places in which Porphyry justifies Aristotle's order, his examples of qmlia are just four: hot, cold, dry, moist.63 This sug-gests that somewhere at the back of his mind the four elements and the four primary qualities are rumbling. The suspicion receives con-firmation from Simplicius, who quotes Porphyry as follows:64

ό δε Πορφύριος παραστήναι τη κατά τον 'Αριστοτέλη τάξει βουλόμενος έκ συνόδου φησιν τοϋ ποσοϋ και τοΰ πρός τι την γένεσιν εχειν τό ποιόν, ωσπερ άλλοις τισιν και Έμπεδοκλεΐ δοκεΐ από της έναρμονίου τών στοιχείων

60 Porph., In Cat. 100; 111; 127. " Porph. apud Simpl., In Cat. 48. " Porph., In Cat. 133,14-19. 63 Porph., In Cat. 127,9. " Simpl., In Cat. 158,27-33.



310 S T E N EBBESEN

κράσεως τάς ποιότητας άναφαίνοντι- και Πλάτων δέ δια των ήμιολίων και διπλασίων και έπιτρίτων και έπογδόων και των τοιούτων έπί του ποσοϋ κατά τό πρός τι θεωρου-μένων λόγων τοϋ τε σώματος τό τοιόνδε είδος και της ψυχής συνέστησεν τό κατά τοιάνδε ποιότητα άφωρισ-μένον.

This must come from the commentary Ad Gedalium. Now, ms. Laurentianus 72,15 (13thc.) contains on f. 22 a scholium which I think has not been printed or discussed before. It runs like this:

πολλοί έτέρως πως είπον προταγήναι τά πρός τι τοϋ ποιοΰ. ό δέ 'Αλέξανδρος δεικνύς δτι φυσική τάξει προ-ετάγησαν τοΰτο πιστοί μεν και τον Έμπεδοκλέα παριστών δοξάζοντα πρώτως μεν στοιχεία δ, έν οίς ευθύς τό ποσόν είτα συνερχόμενα και είρηνεύοντα, έν φ ή σχέ-σις έν fj τά πρός τι, τον νοητόν άνελ(θ)εΐν ϊ διάκοσμον νεΐκος δέ πάλιν σχόντα, οπερ κατά τάς ποιότητας έστιν έν φ τό ποιόν, τά ύπό την αΐσθησιν άναφαίνεσθαι1'. ό δέ Πορφύριος την πρώτην και άνείδεον ΰλην φησίν πρώ-τως1 προβήναι είς άποιον σώμα* και συν τούτφ ευθύς άμα τριχη διαστατόν, μήκος βάθος και πλάτος, έν οίς τό ποσόν και πρός τούτοις συμφυώς τό μείζον και τό ελατ-τον, άπερ των πρός τι· και τότε είς ποιότητας·1. a άνελεϊν cod. b έμφαίνεσθαι cod. c vel πρώτον cod. d ποιο' cod.

The scholiast cannot have got this from Simpl., In Cat. The part about Porphyry could be based on his minor comm. Cat., but I have found no possible source for the part about Alexander. The most likely origin of the scholium is a commentary on the Categories, and it looks very much as if Simplicius and the scholiast are actually reporting each one part of a passage from Ad Gedalium in which Porphyry related and discussed Alexander's use of Empedocles to support Aristotle's order of treating the categories.

The syntax of the scholium is not too clear, but the sense is clear enough, at least if my άνελθεΐν (or άποτελεΐν or the like) is accepted for the manuscript's άνελεϊν. The interpretation of the reigns of Love and Strife as the realms of the intelligible and the sen-




sible, respectively, is well attested in late ancient texts.651 am not sure how the interpretation of Empedocles which the scholiast attributes to Alexander can be fitted in with the information on the same mat-ter that can be gathered from Simplicius' commentaries on the Phys-ics and De caelo. Nor am I sure whether Porphyry in Ad Gedalium remained on the superficial level of the minor commentary when talking about matter, elements and qualities.

If the scholium is really independent of previously known texts, it must be used in the reconstruction of both Alexander's and Por-phyry's thought. I hereby submit the scholium and its puzzles to the consideration of more competent scholars.

65 E.g., Simpl., In Phys. 31,18sqq.



kommentierung, Überlieferung, nachleben () || boethius as an aristotelian scholar

Documents