Astronomical Images and Planetary Theory in Carolingian Studies of Martianus Capella

Pedersen O. , “The ecclesiastical calendar and the life of the Church”, in Gregorian reform of the calendar , ed. by Coyne G. Hoskin M. , and Pedersen O. (Vatican City, 1983), 17–74, presents an excellent introduction to the early history of calendar calculation. A careful rereading of Bede's “De temporum ratione”, in Bedae Venerabilis opera, vi: Opera didascalica, 2 (CCSL, 123 B), ed. by Jones C. (Turnhout, 1977), will always reward the reader.

Bede's Ecclesiastical history of the English people , ed. and transl. by Colgrave B. and Mynors R. (Oxford, 1969), 294–309. Interesting and useful commentary on this chapter in Bede by Wallace-Hadrill J. , Bede's Ecclesiastical history of the English people: A historical commentary (Oxford, 1988), 124–30.

All of these had passed out of currency by the latter half of the seventh century with the exception of Pliny's work, of which Book II was known to Bede. See Eastwood B. , “Plinian astronomy in the Middle Ages and Renaissance”, in Science in the early Roman Empire: Pliny the Elder, his sources and influence , ed. by French R. and Greenaway F. (London, 1986), 197–251, pp. 201 and esp. 224–5 nn. 34, 36–38. Borst A. , Das Buch der Naturgeschichte: Plinius und seine Leser im Zeitalter des Pergaments (Heidelberg, 1994), 98–110, argues extensively for Bede's knowledge of much more of Pliny than Books I-VI but does not overcome the arguments already put forth in the pages previously noted here.

The revival of a “liberal arts” curriculum and the integral part of Capella's Marriage of Philology and Mercury in this curriculum of the schools is described by Contreni J. , “The Carolingian Renaissance: Education and literary culture”, in The new Cambridge medieval history, ii: c. 700 — c. 900, ed. by McKitterick R. (Cambridge, 1995), 709–57.

Pliny's availability was assumed by Alcuin in 798, when he wrote to the king and included a request that a copy of the Natural history be sent to him. Again, in 799, Alcuin presumed that Charlemagne had a copy of Pliny to consult. See Monumenta Germaniae historica: Epistolae , iv, ed. by Dümmler E. (Berlin, 1895), 250, 280 (nrs. 155, 170). Calcidius's presence at court is witnessed by Paris, Bibliothèque Nationale de France (hereafter BNF) ms. lat. 2164; see Bischoff B. , Manuscripts and libraries in the Age of Charlemagne, trans. by Gorman M. (Cambridge, 1994), 29, 64, 139. A copy of Macrobius was used by Dungal of St Denis in his letter of 811 to Charlemagne; see MGH: Epp. , iv, 570–8. The dating of the ninth-century manuscripts of Capella is not well developed; no complete text can be proven to date from much before mid-century. For a convenient list of speculative dates and locations for most of the ninth-century copies of De nuptiis, see Préaux J. , “Les Manuscrits principaux du De nuptiis Philologiae et Mercurii de Martianus Capella”, in Lettres latines du Moyen Âge et de la Renaissance , ed. by Cambier G. Deroux C. , and Préaux J. (Collection Latomus, 158; Brussels, 1978), 76–128, esp. pp. 78–79. Préaux noted the possible dating of some mss. well before mid-century, but this does not necessarily pertain to glosses or diagrams. Among the earliest evidence is Paris, BNF, ms. lat. 13955, ff. 46v–53v, which contains a complete and heavily glossed text of Capella's astronomical book. Wesley Stevens offered the opinion in private correspondence (February 1993) that this manuscript, an important text for teaching the quadrivium, was compiled sometime during the interval 825–840.

There has been disagreement about the date of John Scot's commentary. Scottus Iohannes , Annotationes in Marcianum, ed. by Lutz C. (Cambridge, Mass., 1939), 225–6, presented an appendix regarding two dates given in the commentary. The editor concluded that the first dating clause was too imprecise for use and that the second dating clause fell in 819, too early to apply to John Scot, all of which meant that the two dates in the commentary “were selected entirely at random”. Coulter C. , “The date of John the Scot's Annotationes in Marcianum”, Speculum, xvi (1941), 487–8, corrected Lutz's discussion and proposed a reading of the two dating clauses to put the composition of the commentary in the year (beginning on 25 March) 859–860. However, her argument depended upon identifying the bissextilis as 25 February, when it must be 24 February. This error by Coulter destroys the rest of her argument, which requires a Sunday on 25 February; the text places it on 24 February. A third attempt at dating appeared in van de Vyver A. , “Hucbald de Saint-Amand, écolâtre, et l'invention du Nombre d'or”, Mélanges Auguste Pelzer. Recueil de travaux d'histoire et de philologie , 3e sér., xxvi (Louvain, 1947), 61–79, see p. 64, n. 16. By arguments which cannot be validated he concluded that the commentary was done in 852. Referring to the same two dating clauses as Lutz and Coulter, van de Vyver claimed that 12 September in 852 fell on luna xx, which is incorrect. His assumption that xii kalendarum Septembris may stand for 12 September was also mistaken. Other errors make his argument confusing and useless. All attempts at finding a date of composition on the basis of the two dating clauses have failed. Lacking other interpretations, it seems that these two passages were composed either at different times or without intent to correspond to an actual date. H. Liebeschütz proposed that the Paris ms. containing the commentary on all books of De nuptiis should be dated after the predestination controversy of 851. See his “The place of the Martianus Glossae in the development of Eriugena's thought”, The mind of Eriugena , ed. by O'Meara J. and Bieler L. (Dublin, 1973), 49–58, esp. p. 53. This view, based on contextual interpretation of two crucial manuscripts, would seem to be the most authoritative and the most useful. I am deeply indebted to Professors D. O Cróinín and W. Stevens for their separate critical analyses (in private correspondence) of the dating clauses and the arguments of Coulter and van de Vyver.

For these manuscripts see the thorough catalogue of Leonardi C. , “I codici di Marziano Capella”, Aevum, xxxiii (1959), 443–89; xxxiv (1960), 1–99, 411–524. See catalogue entries, nrs. 8, 20, 26, 28, 37–38, 73, 81–84, 89, 101, 144, 160–2, 171, 208, 210.

Lutz C. , “Martianus Capella”, Catalogus translationum et commentariorum, ii (1971), 370–2, for three pre-Remigian commentaries. Lutz C. and Contreni J. , “Martianus Capella, addenda et corrigenda”, ibid., iii (1976), 449–52, pp. 451–2.

Lutz , “Martianus Capella”, 372–6. See Autissiodorensis Remigius , Commentum in Martianum Capellam, ed. by Lutz C. (Leiden, 1962–65). Further regarding Capellan glosses, see Leonardi C. , “Remigio d'Auxerre e l'eredità. della scuola carolingia”, I classici nel medioevo e nell'umanesimo: Miscellanea filologica (Genoa, 1975), 271–88; idem, “Glosse eriugeniane a Marziano Capella in un codice leidense”, Jean Scot Erigène et l'histoire de la philosophie (Laon, 7–12 juillet 1975) (Paris, 1977), 171–82.

Leonardi C. , “Raterio e Marziano Capella”, Italia medioevale e umanistica, ii (1959), 73–102. Leonardi speculatively dated the ms. to mid-century or about 847. I find it sensible to assume a date of composition before mid-century without speculating how long before. See the description and bibliography in de Meyier K. , Codices Vossiani latini, i: Codices in folio (Leiden, 1973), 100–3.

For clarity I identify now the group deriving from Leiden, Voss. F.48, based on the diagram tradition, as Leiden, UB, ms. B.P.L. 36; Paris, BNF, ms. lat. 8671; and Paris, BNF, ms. lat. 8669 (the last ms. has an unlabelled copy of the diagrams). It is interesting that two of these three later sets of diagrams may have been done at Auxerre, since Voss. F.48 came to the Auxerre library quite early; both Paris 8671 and Leiden BPL 36 have been speculatively said by Préaux to have originated at Auxerre. The third ms., Paris 8669, has been tentatively linked to Soissons. See Préaux , “Manuscrits principaux” (ref. 5), 79. Bischoff B. , Lorsch im Spiegel seiner Handschriften (Munich, 1974), 45, 96–97, said that Leiden B.P.L. 36 was written at Lorsch.

This group is Leiden, UB, ms. B.P.L. 87; Leiden, UB, ms. B.P.L. 88; and Vat., ms. Regin. lat. 1987.

Besançon , Bibl. Mun., ms. 394.

Detailed discussion of the text and its history from the Carolingian era forward appears in Eastwood B. , Astronomy and optics from Pliny to Descartes (London, 1989), chap. 2. Whether or not the idea was independently available in other Latin texts during the Middle Ages and known to Copernicus is discussed in detail in chap. 1 of the same book.

The translation I give here is essentially the same as found in Martianus Capella and the seven Liberal Arts, ii: The marriage of Philology and Mercury, transl. by Stahl W. and Johnson R. (New York, 1977), 332, 333 (sects. 854, 857), except for the last line. There Stahl has “broader” where I have “narrowly restricted”. Stahl depended upon the 1925 edition of Adolf Dick, whose reading here was also followed in the latest edition, Martianus Capella, ed. by Willis J. (Leipzig, 1983), 324.17 [hereafter: Ed. Willis]. While there are some problems with the Latin text, it is quite clear that every medieval reader who inspected and glossed this text discerned the meaning I give in my translation, not the meaning given by Stahl's translation. At the point where the maximum elongation of the two planets is stated, I follow an emendation found in many medieval readings as well as in modern editions; this does not affect the point of debate in the subsequent sentences. Fuller discussion of ms. readings at this point appears in Eastwood , Astronomy and optics (ref. 14), chap. 2, pp. 146–55. The Latin text of the Vossius, which may be compared with the modern edition, reads as follows: “… tria item ex his cum sole lunaque orbem terrae circumeunt, Venus vero ac Mercurius non ambiunt terram.… nam Venus Mercuriusque licet ortus occasusque cotidianos Ostendant, tamen eorum circuli terras omnino non ambiunt, sed circa solem laxiore ambitu circulantur. Denique singulorum suorum centron in sole constituunt, ita ut supra ipsum aliquando, intra plerumque, propinquiores terris ferantur; a quo quidem uno signo et parte dimidia Mercurius disparatur. Sed cum supra solem sunt, propinquior terris est Mercurius; cum intra solem Venus, utpote quae orbe castioque diffusioreque curvetur.” Leiden, UB, ms. Voss. F.48, f. 79r, 29 and 79v, 9–11.

Beside the concentric model, this explanation is given: “Haec figura pertinet ad hoc quod dicit quod aliquando cum infra solem feruntur propinquiores terre sunt, et ad illud etiam quod dicit quia cum superiores sunt, propinquior est Mercurius terre.”

For the modern edition see Martianus Capella, ed. Willis , 324.16–17; note also the apparatus to line 17 for actual mss. readings. For the Vossius text, see ref. 15 above.

The comment reads as follows in the outer margin of f. 79v of Voss. F.48. “Si secundum Platonem ordinem planetarum voluit ostendere, in hac sententia terris potest stare, si vero secundum Pitagoricos Pliniumque ordinem planetarum velimas [sic] sumere, nunquam intellegere poterimus nisi ablatum fuerit terris ut sic sententia scribatur, sed cum supra solem sunt propinquior Mercurius est ut subaudiatur soli.”

In the following description from Pliny I emphasize the words that would be especially suggestive for the Plinian model drawn in the Vossius. “… cur Veneris Stella numquam longius XLVI partibus, Mercurii XX ab sole abscedant, saepe citra eas ad solem reciprocent. Conversas habent utraque apsidas ut infra solem sitae, tantumque circulis earum subter est quantum superne praedictarum, et ideo non possunt abesse amplius, quoniam curvatura apsidum ibi non habet longitudinem maiorem, ergo utrique simili ratione modum statuunt apsidum suarum margines, ac spatia longitudinis latitudinum evagatione pensant. At enim cur non semper ad quadraginta sex et ad partes viginti perveniunt? Immo vero, sed rationes canonicos fallit. Namque apparet apsidas quoque earum moveri, quod numquam transeant solem.” Secundus Plinius C. , Naturalis historia libri XXXVII, ed. by Jan L. and Mayhoff K. (Leipzig, 1906), i, 149.25–150.12 (Book II, 72–73).

On Dungal's letter see Eastwood B. , “The astronomy of Macrobius in Carolingian Europe: Dungal's letter of 811 to Charles the Great”, Early medieval Europe, iii (1994), 117–34. For the use of Plinian astronomy in the computistical compilations coming from the Computus of 809 see Eastwood B. , “Plinian astronomical diagrams in the early Middle Ages”, in Mathematics and its applications to science and natural philosophy in the Middle Ages, ed. by Grant E. and Murdoch J. (Cambridge, 1987), 141–72.

See for example Eastwood , “Plinian astronomy” (ref. 3), 198–9, 207–9.

Plato Latinus, iv: Timaeus a Calcidio translatus commentarioque instructus, ed. by Waszink J. (London, 1962/1975), 148.20–150.6 (c. 97).

Ibid. , 158.9–159.13 (c. 112). The larger context of this passage, its part in the history of doctrines of early Greek heliocentrism, and a medieval reconstruction of the diagrams for bounded elongation to accompany the text are all discussed by Eastwood B. , “Heraclides and heliocentrism: Texts, diagrams and interpretations”, Journal for the history of astronomy, xxiii (1992), 233–60.

This text, taken from the preface to Pseudo-Cyprian's De duodecim abusivis saeculi, was probably inserted after the addition of ff. 93–94 to Capella's work. These subsequent folia contain a commentary of prophetic interest by Pseudo-Jerome, which can have inspired the brief addition from Pseudo-Cyprian of related concern, inserted in the centre of the preceding page. More speculatively, one might see this insertion as an admonition regarding excessive attention to knowledge of the purely physical world, viz., detailed astronomical knowledge. On the texts here of Pseudo-Cyprian and Pseudo-Jerome, see de Meyier , Codices Vossiani latini (ref. 10), i, 102. For the influence of Pseudo-Cyprian's work, see Anton H. , “Pseudo-Cyprian, De duodecim abusivis saeculi und sein Einfluss auf den Kontinent, insbesondere auf die karolingischen Fürstenspiegel”, Die Iren und Europa im früheren Mittelalter, ed. by Löwe H. (Stuttgart, 1982), ii, 568–617; see pp. 604, n. 115 and 605, n. 123 for the Vossius.

This figure referred to a passage in sect. 867 (ed. Willis , 328.14–16), where Capella wrote of the Sun's latitudinal motion, “Sol enim in nullam excedens partem medio libramento fertur absque ipso Librae confinio; nam ibi se in austrum aquilonemve deflectit ad dimidium fere momentum.” I have emphasized the phrase used by the glossator for a reference marker.

Solar and lunar eclipses were described by Capella at sects. 869–70 (ed. Willis , 329.7–18).

At sect. 868 (ed. Willis , 328.19–22) Martianus described the oblique path of the Moon across the zodiac. The central part of the passage, with reference phrase emphasized here, is: “ita tamen ut descendens ascendensque ipsam solarem lineam, quam mediam inter senas utriusque lateris partes esse monstravi, aut acutis aut spatiosis angulis secet” (lines 20–22).

Martianus described the relative lengths of the planetary orbits, emphasizing the increase in length passing from the centre to the periphery, in sect. 861 (ed. Willis , 326.7–15).

This somewhat obscure image, clarified by its labelling phrase, illustrated sect. 880 of Capella's astronomy. The reference phrase appears in the following sentence: “Quippe ubi radiis solaribus condicione partium liberatus ante emergentis splendorem iubaris vibrabundus apparet” (ed. Willis , 333.19–20).

This figure offers an intriguing look into the glossator's attempt to explain a corrupt text with no apparent suspicion that a corruption existed. The image illustrated the same general passage, regarding the motion of Mercury (sect. 880), as the previous image. In this case an earlier part of the passage stated: “sed idem Stilbon, licet solem ex diversis circulis continetur, ab eo tamen numquam ultra xxii partes poterit aberrare nec duobus signis absistere …” (ed. Willis , 333.14–16). A modern scholar might easily assume that the difference between the modern, critical edition and the particular diagram in the Vossius was a copyist's slip, converting ‘xxii’ into ‘xxx’. But this was not the case. First, although available authorities such as Pliny informed any student of astronomy that Mercury's maximum elongation from the Sun was 22° or very close thereto, the relevant text of the Vossius reads: “ultra xxxii partes” (f. 81v, 13). An interlinear gloss adds “sub sole” in order to emphasize that the number of degrees does not refer to bounded elongation, measured along the Sun's path, but rather to a quantity measured below the Sun's orbit. Guaranteeing that we are not witnessing one aberrant reading and one aberrant gloss (and diagram), are the following facts. Almost all of the mss. consulted by Willis for his edition have the presumably erroneous number (ed. Willis, 333, apparatus to line 16). The same diagram with label appears in many ninth-century manuscripts. The text, diagram, and presumed explanation were sufficiently widespread to find approval and inclusion in the commentary on Capella by Remigius of Auxerre. See Remigius , Commentum, ed. by Lutz , ii, 290 (at 464.10) and again ii, 291 (at 465.10); the latter of these Remigian glosses makes very clear that the “xxxii partes” was not considered the standard elongation.

This figure illustrated sect. 872, ed. Willis , 330.9–22, esp. at lines 16–17: “nam primus Arietis circulus primus est Librae, item secundus ac tricesimus.”

Here the image illustrated sect. 875, where Capella began to describe the varying lengths of daylight and nighttime. Capella then proceeded to discuss varying measures of the extremes at different latitudes. See ed. Willis , 331.19–333.2 (sects. 875–7).

At sect. 844 Capella began a lengthy discussion of the varying speeds for risings and settings of the zodiacal signs, with oblique (“inequales”) risings faster than vertical (“equales”) risings and the same for settings. See ed. Willis , 319.3–6, where the two terms in the text are “transversa” and “recta” for “inequales” and “equales” in the diagram.

This manuscript is Paris, BNF, ms. lat. 8669, the diagrams on f. 122r-v.

Paris, BNF, ms. lat. 8671, f. 84r.

Leiden, UB, ms. B.P.L. 36, f. 129r.

This additional and unlabelled diagram did, however, show up again in a set of astronomical diagrams appended to Remigius's commentary on De nuptiis, in Paris, BNF, ms. n.a.l. 340, f. 83r (s. X), where eight of the ten astronomical diagrams, including the new one, in this set were copied. However, the three-version diagram carries no labels of attribution, and the copyist reproduced not only the combination diagram but also each of the three versions separately on this page.

A date of 836–848 was set for the Bedan texts in Karlsruhe, LB, ms. Aug. CLXVII, according to the catalogue of Holder A. , Die Handschriften der … Landesbibliothek in Karlsruhe, v: Die Reichenuaer Handschriften, I. Die Pergamenthandschriften (Karlsruhe, 1906), 396. The manuscript was discussed in Bedae Opera de temporibus , ed. by Jones C. (Cambridge, Mass., 1943), 145–6, where its earliest possible date was set at 822. More recent discussion appears in Bischoff B. , “Irische Schreiber im Karolingerreich”, in idem. Mittelalterliche Studien: Ausgewählte Aufsätze (Stuttgart, 1981), iii, 48–49.

Jones C. , “A note on concepts of the inferior planets in the early Middle Ages”, Isis, xxiv (1936), 397–9, reproduced the images on f. 16r of the Reichenau ms., pointing out their connection to Martianus Capella's doctrine on the inner planets. The diagrams appear on p. 398 of Jones's Note.

Folio 18bis verso: “Nam Venus Mercuriusque licet ortus occassusque cotidianos ostendant tamen eorum circuli terras omnino non ambiunt sed circa solem laxiore ambitu circulantur, denique circulorum suorum centron in sole constituuntur, ita ut supra ipsum aliquando intra plerunque propinquiores terris feruntur, a quo quidem uno signo et parte dimedia Venus disparetur, sed cum supra solem sunt propinquior est terris Mercurius, cum intra solem Venus ut pute que orbe castiore diffusioreque curventur.…” Compare the Capellan text in ed. Willis , 324.10–17.

The gloss is to “Venus” at f. 27vb, 10. The gloss itself appears in the upper margin of column a and reads: “Venus et Mercurius. Eorum vero circuli terras omnino non ambiunt sed circa solem laxiore ambitu circulantur. Denique circulorum suorum centron in sole constituunt.” Comparison of this gloss with the longer one on f. 18bis verso shows that this one on f. 27v is exactly the same, verbatim, as a section of the previous gloss, viz., “eorum circuli … in sole constituuntur”. See ref. 40 above.

Examples are (1) Melk, Stiftsbibl., 412 (370. G 32), p. 35; (2) Berlin, SB, Phillipps 1832, f. 22r; (3) Paris, BNF, lat. 5239, f. 198r; (4) Vat., Ross. lat. 247, f. 83r; (5) Wolfenbüttel, HAB, 127 Gud. lat. folio, f. 10v. Some of these are discussed by Jones C. , “The Byrhtferth glosses”, Medium aevum, vii (1938), 81–97. Contreni J. , The Cathedral School of Laon from 850 to 930, its manuscripts and masters (Munich, 1978), 124–8, 150–1, discusses Phillipps 1832 and Melk 412.

Florence, Bibl. Ricc., ms. 916, f. 88r, and Oxford, Merton Coll., ms. 291, f. 94v.

Paris, BNF, ms. lat. 14754, f. 188r.

Naples, BN, ms. V.A.16, f. 228v (s. XV).

An early example: Munich, SB, clm 14729, f. 221v (s. X). Two late examples: Florence, BL, ms. P1. 51.13, f. 127v (s. XV); Vat., Urb. lat. 329, f. 139v (s. XV).

Leiden, UB, B.P.L. 87; see Préaux , “Manuscrits principaux” (ref. 5), 79.

On f. 126r is “aut acutis aut spatiosis” in the same form as found in the Vossius. On f. 126v is “Libra, Aries” in the same form as the Vossius shows.

These are Leiden, UB, B.P.L. 88 and Vat., Regin. lat. 1987.

In B.P.L. 88 on f. 164r, in Regin. 1987 on f. 129v. This way of illustrating the matter is more cumbersome and potentially ambiguous.

In B.P.L. 88 on f. 163v; in Regin. 1987 on 129r.

Figures 9 and 10 show the diagrams in B.P.L. 88; the diagrams in Regin. 1987 are exactly the same as those in the Leiden ms.

In B.P.L. 88 the concentric model on f. 162 and the intersecting circles on f. 162v are labelled by the same hand, which is quite distinct from the hand of the interlinear glosses and the marginal comment. In Regin. 1987 the concentric model on f. 127v and the intersecting circles on f. 128r are likewise labelled by the same hand. In B.P.L. 88 the “terris” in the text is underlined and “id est soli” appears in an interlinear gloss. In Regin. 1987 the text has a lacuna of one line at this point (line 12), which is filled by a marginal gloss, and in this gloss the “terris” is omitted and “ei” is inserted in its place.

Besançon , Bibl. Mun., ms. 594, ff. 85. See Préaux , “Manuscrits principaux” (ref. 5), 79; Leonardi , “Codici” (ref. 7), 13–14 (#20).

The comment appears in the inner margin of f. 71r.

This appears in the inner margin of f. 72r.

This appears in the outer margin of f. 72v.

In Voss. F.48, ff. 80v–81r.

These are Bamberg, SB, Class. 39; Oxford, Bodl., Laud lat. 118; and Paris, BNF, lat. 8670.