Vous êtes sur la page 1sur 422

In rem em brance

A L B E R T O C O FFA (1935-1984)
Professor of History and Philosophy of Science
Indiana University-Bloomington
WILLIAM C O L E M A N (1934-1988)
Dickson-Bascom Professor of Humanities and
Professor of History of Science and Medicine
University of Wisconsin-Madison
V I C T O R E. T H O R E N ( 1 9 3 5 - 1 9 9 1 )
Professor of History and Philosophy of Science
Published by the Press Syndicate of the University of Cambridge Indiana University-Bloomington
The Pitt Building, Trurnpington Street, Cambridge CB2 IRP
40 West 20th Street, New York, NY 10011-4211, USA
10 Stamford Road, Oakleigh, Melbourne 3166, Australia
© Cambridge University Press 1996
First published 1994
Reprinted 1996
First paperback edition 1996
Printed in the United States of America
Library o f Congress Cataloging-in-Publication Data is available.

A catalog record for this book is available from the British Library.
ISBN 0-521-43344-4 hardback
ISBN 0-521-56509-X paperback
12 ON CELESTIAL MATTER 245

During the Middle Ages, all were agreed with Aristotle and his com­
mentator, Averroes, that generation and corruption were processes asso­
ciated inherently with sublunar matter compounded o f the four elements.
O n celestial matter: Can it O nly the latter, and the bodies compounded o f them, could undergo change
from contrary qualities. W ithout contrary qualities, the celestial ether could
exist in a changeless state? not change and must therefore be incorruptible.
D id it follow that matter was something associated solely with terrestrial
bodies, whose never-ending succession o f contrary qualities produced in­
cessant change? D id the absence o f such contrary qualities in the heavens
also im ply an absence o f matter as well? O r did the heavens include matter
o f some kind, and were they composed o f matter and form? O n this issue,
That medieval scholastics regularly wondered whether the heavens, with like many others, Aristotle provided no clear guidance. Indeed, passages
all o f the planets and stars, possessed matter may at first glance appear from his many works could readily be selected both to support and to
strange or even starding. Because the planets and stars are readily visible, oppose the idea o f a celestial matter / The medieval controversy over celestial
and because visible effects were associated w ith matter,1 it should have been matter involved tw o issues. The first and prior issue pitted those who denied
obvious that some klhd o f matter must underlie the celestial appearances. its existence against those w ho affirmed it. Because most affirmed the ex­
Although most scholastics accepted the existence o f such matter, others istence o f some kind o f matter in the heavens, the secondary and more
found it contrary to the principles o f natural philosophy. widely debated dispute concerned the nature and properties o f an unchang­
The question as to whether matter existed in the celestial region was ing matter and the manner o f its existence.5
thrust upon the Middle Ages by Aristotle’s division o f the w orld into
radically different celestial and terrestrial (sublunar) regions. In Chapter io,
which concerned celestial incorruptibility, an issue that is intimately con­
nected to the problem o f celestial matter, w e noted Aristotle’s justification
for this division: the existence o f tw o radically different kinds o f motion in I. That matter does not exist in the heavens
the universe. One - finite, rectilinear, and therefore incomplete - was as­ As Buridan and others were well aware, “ philosophers are accustomed to
sociated with elemental bodies and bodies compounded o f those four ele­ use the term ‘matter’ in many w ays,” 6 some o f which clearly applied to the
ments; the other - circular and complete, without beginning or end - was heavens. Thus, i f matter is conceived as something composed o f quantitative
the motion o f celestial bodies alone. It followed for Aristotle that celestial parts, the heavens must obviously possess matter, since they are composed
bodies and the heavens in general consisted o f a substance - or ether - o f quantitative parts. The heavens would also possess matter i f matter were
different from any o f the four elements. Incorruptibility, and therefore defined as a substance that is the subject o f motion or other accidents.7 But
unchangeability, were its most fundamental properties, properties so amaz­ these were not the senses in which those who rejected the existence o f
ing that they made the celestial substance radically different from bodies in celestial matter - for example, Buridan, Godfrey o f Fontaines, Peter Aureoli,
the terrestrial region.2
We have also examined the theoretical basis for celestial incorruptibility:
the absence o f contrary qualities in the celestial region. In Galileo’s words, 4. For an excellent illustration o f h o w such juxtapositions were form ed, see ibid., qu. 5 (K),
1 0 3 - m (for the w a y in w hich A verroes cited A ristotle’s w orks against the existence o f
“ Whatever undergoes corruption has a contrary. . . ; and therefore. . . the celestial matter); m - 1 1 2 (for those w h o cited A ristotle in favor o f celestial matter); and
heavens, since they lack contraries o f this type, are incorruptible.” 3 More­ 1 1 2 - 1 1 5 (for citations concerning the idea that celestial matter differs from, or is identical
over, Aristotle buttressed theory with the empirical claim that no changes to, terrestrial matter). Inspection o f similar questions b y numerous other scholastic authors
w ou ld reveal the same pattern.
in the heavens had ever been detected or recorded. 5. In w hat follow s, I rely heavily on Grant, 1983.
6. “ N otan d um est quod multis solent philosophi uti hoc nom ine ‘materia.’ ” Buridan [D e
caelo, bk. 1, qu. 11], 1942, 49. O resm e [D e celo, bk. 1, qu. 11], 1965, 160-162, gives four
1. In Metaphysics 8.1.10423.24—26, Aristotle says that “ sensible substances all have matter.”
different senses o f “ m atter” and remarks that Aristotle speaks o f matter in a variety o f
See Aristotle [Ross], 1984.
w ays in the second bo ok o f his Physics.
2. In Metaphysics 12.7.10743.30—37, Aristotle em phatically denies that m atter can exist in
7. Buridan, ibid. For the locus o f these ideas, Buridan cites A ristotle’s Metaphysics, bks. 7
the heavens. For a brief discussion, see the beginning o f Chapter 12 and C hapter 12, Sec­
and 8, respectively. A lbert o f S axon y [D e celo, bk. 1, qu. 4], 1518, 89r, col. 1, w hose text
tion I.
is occasionally an almost verbatim c o p y o f Buridan’s D e caelo, gives virtually the same
3. Galileo [D e caelo, qu. 4 (J)], 1977, 98.
references to Aristotle.

244
246 THE CELESTIAL REGIO N ON CELESTIAL M ATTER 247

and Albert o f Saxony - considered the question.8 As Buridan explained, Peter Aureoli, a student o f Duns Scotus and a trained theologian, had
for them “ matter is called that from which a substance is composed with arrived at similar ideas in a much earlier discussion o f the subject in his
a substantial form inhering in it, which persists by itself, and which is called Commentary on the Sentences, where he also considered whether “ the au­
‘this something.’ ” 9 thorities o f Sacred Scripture and the Catholic Doctors” thought it objec­
B y the time Buridan wrote, the opinions o f Thomas Aquinas and Ae- tionable to assume that the heavens consist o f a simple, incomposite
gidius Romanus (which are described in Section II o f this chapter) were substance.14
taken as representative o f tw o rival theories in favor o f the existence o f Aureoli first describes arguments he attributes to Aristotle and Averroes.
celestial matter. Buridan considered and rejected both10*but admitted that These include their hostility to the conception o f the heavens as a composite
it was difficult to refute the claim for the existence o f celestial matter by o f matter and form and their conviction that the heavens have only fixed
demonstrating its opposite: namely, that matter does not exist in the heav­ dimensions that belong to them as an inherent property.15 He then presents
ens. Buridan agreed with Aristotle that something could not be made nat­ his ow n opinions, o f which the most striking is that the heavens not only
urally from nothing. Consequently, the only proper w ay to define matter consist o f a simple, incomposite substance, but that this substance is neither
was by means o f substantial transformations in which something, namely matter nor form .16 Aureoli also rejects the idea o f the heavens as a form,
matter, remained constant as a body was generated and corrupted, that is, or having a form, because form is something that determines and controls
as the same matter acquired and lost successive forms. But like all his fellow matter, accidents, and properties, and the heavens lack any mechanism with
scholastics, Buridan believed that such changes could not occur naturally which to control their properties and perfections.17 The explanation for this
in the celestial region, so that “ it is absolutely in vain and without cogent major difference between the terrestrial and celestial regions lies in the
reason that we should posit such matter in the heavens.” " variability in size o f animate and inanimate things in the former and, by
Admitting, however, that he could not devise a formal demonstration contrast, the invariability o f the figure and size o f bodies in the latter. In
for the nonexistence o f celestial matter, Buridan sought to achieve the same animate, terrestrial bodies, sizes vary within certain limits that are controlled
result by invoking the widely used principles that nature does nothing in by the soul; in inanimate bodies, variations in the size o f a thing are con-
vain and that it is useless to “ save the phenomena” with more when it can
(and then rejected) the opinion Buridan favored w hen he declared that supporters o f this
be done with less.12 Rather than assume the existence o f celestial matter, opinion say that “ there is no potentiality for substantial being, nam ely for a substantial
Buridan believed that all the phenomena could be saved by the assumption form . A n d this is the position o f the Com m en tator [i.e. Averroes] in the D e substantia
orbis and o f certain m odem s w h o say that the heaven is a certain form spread out and
o f a simple, uncomposed, celestial substance which, because it functions as
extended b y quantity, but w hich is supported b y nothing” (Hervaeus Natalis, D e materia
a subject for an extended magnitude, must possess the property o f extension. celi, qu. 3, 1513, 38V, col. 2). Th is opinion was also reported b y Peter o f A ban o in his
Indeed, this same celestial substance also serves as a subject for motion and Lucidator dubitabilium astronomiae, differ. 5, 1988, 288. A s Peter puts it, the planets and
stars are forms or dimensions that are not in matter, but are “ spiritual bodies” (corpora
other accidents.13
spiritualia) or can be said to possess matter equivocally. Averroes had assumed indeter­
minate dimensions as an inherent property o f the celestial bo dy (see Averroes, D e substantia
8. Galileo adds names to the list o f those w h o rejected celestial matter, including Durandus orbis [H ym an and Walsh], 1973, 3x2). These dimensions, w hich were associated w ith the
de Sancto Porciano, Duns Scotus, Marsilius o f Inghen, John o f Jandun, “ and all A v e r - celestial ether’s substantial form , w ere not divisible into determinate quantities, as were
roists.” Galileo, D e caelo, qu. 5 (K), 1977, 105. the indeterminate dimensions associated w ith prime matter in the terrestrial region.
9. “ Sed proprie loquendo materia vocatur ex qua cum forma substantiali sibi inhaerente 14. “ Articulus II: U trum ponere caelum et naturam simplicem et non com positam ex materia
com ponitur substantia per se subsistens quae dicitur ‘hoc aliquid.’ ” Buridan, D e caelo, et forma repugnat auctoritatibus sacrae Scripturae et D octorum C ath olico ru m .” Aureoli
bk. 1, qu. 11. I 94 2> 49- Hervaeus Natalis, D e materia celi, qu. 3 (“ Queritur utrum corpora [Sentences, bk. 2, dist. 14, qu. 1, art. 2], 1595-1605, 2:189 col. 1.
omnia superiora et inferiora com m unicent in m ateria” ), 1513, 38V, col. 1, put it in much 15. Ibid., 186-87. If the dimensions o f the heavens were as indeterminate as prime matter,
the same w ay when he declared that speaking “ m ore strictly, matter can be taken in the heavens w ould necessarily change their dimensions in the same manner as terrestrial
another w ay, namely, as that w hich underlies a substantial form and is part o f a com posite bodies. Thus if the heavens were to be assumed incorruptible, Aristotle and Averroes,
subject. A n d n ow w e shall speak about matter in this manner. ” (A lio m odo potest accipi and all w h o follow ed them, had to assume invariant celestial dimensions. T h e heavens,
materia magis stricte illud, scilicet quod substem itur form e substantiali et est pars sub- or any part o f them, are therefore not capable o f increasing or decreasing their size b y
stantie composite. Et sic nunc loquim ur de materia.) rarefaction or condensation or b y addition or subtraction o f any im aginable substance.
10. Buridan, D e caelo, bk. 1, qu. 11, 1942, 5 1 -5 2 . Albert o f Saxon y, D e celo, bk. 1, qu. 4, See Aureoli, ibid., 187 col. 1 ( D -F ). O n the difference between determinate and inde­
1518, 89r, cols. 1 -2 , also rejected the same tw o theories. terminate dimensions or quantities, see Weisheipl [M cM ullin], 1963, 1 4 7-16 9 .
xi. “ ldeo frustra om nino et sine ratione cogente ponerem us talem materiam in caelo.” B ur­ 16. “ C aelum esse quantum non com positum , scilicet ex materia et forma, nec forma, nec
idan, ibid., 53. materia, subiectum habens dimensiones tantum in actu su o.” Aureoli, ibid., 189, col. 1
12. Ibid., 52. Buridan attributed these ideas to Averroes, but the latter mentions on ly that (A)- , . , .
nature does nothing in vain (see Averroes [D e caelo, bk. 1, com m ent. 20], 15 6 2 -15 74 , 17. “ D e natura enim formae est quod sit in actu et determinet materiam et largiatur per
5: 1 5r, col. 2). Saving the phenomena w ith the fewest possible assumptions is more akin m odum exigentis et determinantis accidentia et proprietates. . . . Sed natura caeli non
to the principle o f O ck h am ’s razor. determinat proprietates suas et perfectiones postremas. Natura enim caeli in quantum
13 - Buridan, ibid. $2. Hervaeus Natalis, w h o w rote som e years before Buridan, described huiusm odi non determinat sibi tantam quantitatem .” Ibid., 188, col. 1 (E -F ).
248 THE CELESTIAL REGION ON CELESTIAL MATTER 249

trolled by its form. If a cow, for example, were made as long as a serpent, With most o f the traditional features o f the celestial region rejected and
it would cease to be a co w ,18 a possible catastrophe that is avoided by its virtually no really new ones available to replace them, Aureoli was left with
soul. Similarly, if a proper upward motion were absent in fire, the latter an extended magnitude possessed o f certain vital properties conferred by
w ould lack a form . 19 an otherwise undescribed intelligence. B y offering so little information,
B y contrast, celestial matter and its bodies are invariant in size, shape, Aureoli could present the positive features o f his conception o f the heavens
and motion and therefore require no forms or souls to regulate and deter­ in but a few lines.
mine those properties. Indeed, nothing inheres within celestial bodies that Buridan, Aureoli, and others who rejected the existence o f matter in the
would enable them to determine their ow n properties and operations. Con­ heavens were ultimately supporters o f Averroes’ position. And yet Aver­
sequently, the heavens cannot possess a form, or form s.20 As i f to reinforce roes, despite his rejection o f the idea that the celestial substance was a
his argument, Aureoli observes that although the heavens are a finite entity, composite o f matter and form ,23 and his insistence that the celestial substance
they are also eternal and permanent. But a finite thing lacks the nature to had to be something simple and uncomposed, was prepared on occasion
determine an eternally permanent thing. N o intrinsic principle o f the ce­ not only to call that simple substance a form, but also to call it matter,
lestial substance - that is, no form or matter or combination th e re o f- could though clearly not matter in the ordinary sense, as described earlier by
have conferred such properties on the heavens. O nly an external power Buridan. In his commentary on De caelo, Averroes first concludes that “ the
could have bestowed them, a power which Aureoli identifies w ith the celestial body does not have matter,” 24 arguing that even if matter existed
celestial intelligences.'21 in the incorruptible celestial region it would be superfluous, because it could
If the heavens lack both form and matter, what kind o f an entity could never receive a new form (since there are no contrary forms) and thus could
they be? O n this crucial issue, Aureoli, w ho preceded Buridan, was some­ never change. Its potentiality would be forever frustrated and in vain, which
what more forthcoming. He judged the heavens to be an existing magnitude: is contrary to nature.
a quanta esse, as he described it. “Just as matter is not understood except in But later in the same commentary, Averroes speaks o f the “ matter o f the
relation to form ,” Aureoli explained, “ so w e cannot understand that the form o f a celestial body, which [matter] is actualized. ” 2S As “ actualized”
heavens have a definite [or fixed] quantity, figure, motion, and other prop­ matter, it obviously could not be conceived as the prime matter underlying
erties unless [these properties] are [understood in] relation to an intelli­ the four elements, which is a pure potentiality. Although celestial matter
gence.” 22 The quanta esse, or celestial magnitude, functions like a subject lacked any potentiality for substantial change (i.e., generation and corrup­
but has received its properties from an intelligence. W ithout matter and tion), or for qualitative or quantitative change and could not be stripped o f
form, the tw o fundamental principles o f all terrestrial things, the quanta esse its form, it did have a potentiality for place, as the celestial motions made
and its properties fitted none o f the traditional descriptions o f the heavens. evident.26 Thus, with the celestial bodies clearly in mind, Averroes declared,
in his commentary on the Metaphysics, that “ eternal things, which are not
x8. “ U n d e anima quaelibet determinat figuram certam sui corporis sine qua non potest esse:
generable but are moved with a translatory motion, have matter; not, how ­
facias enim bo vem longum sicut serpentem, statim amittet esse b o v is .” Ibid., 188 col. 2
(A). ever, generable matter, but [only] the matter o f those things that are moved
19. “ Si ergo ignis non haberet ex se m otum proprium eum , qui est sursum, qui est ei proprius, from place to place.” 27
iam sequitur quod ignis non est form a.” Ibid., (B).
20. “ Sic in proposito cum caelum sit determinatae quandtatis in actu quia in eo non sunt
dimensiones interminatae, cum etiam sit figurae rotundae et habeat m otum circularem 23. In his Commentary on D e caelo, Averroes declared that the heavens are not com posed o f
sine quibus im po ssible est esse et talia non determinet sibi per naturam propriam, cir­ matter and form as are the four simple elements, because “ forms that are in matter are
cumscripta amma. Patet quod caelum non est forma, cum forma quaelibet se ipsa, omni contraries, and i f a form existed in matter w ithout a contrary, then nature w ou ld act in
alio circumscripto, determinet sibi suas proprietates et operadones. Hanc radonem tangit vain, since no potentiality w hatever could exist in this matter because potentiality occurs
Com m en tator D e substantia orbis, tractstus 2 .” Ibid., ( B - C ) . on ly w hen a form can separate from [its] m atter.” See A verroes, D e caelo, bk. 1, com m ent.
21. “ H o c idem potest apparere de aetema eius permanentia quam sibi non determinat, ut talis 20, 15 6 2 -15 74 , 5:15^, col. 1.
natura est. N ulla enim natura, quae habet dimensiones finitas, videtur sibi ex se deter- 24. Ibid., bk. 1, com m ent. 21, 5:15V col. 2.
minare permanendam aetemam. Sed hoc habet ex determinatione extrinseca, scilicet 25. “ D icam us ergo quod ista natura neutra est, et non existens per se in actu, sed est materia
intelligendae, quae largitur ei omnes huiusm odi perfecdones consequentes, non effective, formae corporis celestis, que est in actu.” Ibid. bk. 1, com m ent. 95, 5:63V col. 2. Here
sed exigid ve et determinative tantum m odo, ut dictum est saepe.” Ibid., ( C - D ) . A verroes speaks o f matter and form.
22. N atura caeli est natura subiecd, et est esse quanta. U n de sicut materia non intelligitur 26. “ E t ideo in hac nulla potentia est qua denudari possit a sua forma et non habet nisi
nisi in analogia ad form am , sic non possumus intelligerc caelum habere determinatam potenriam ad u b i.” Ibid. 5:64r, col. 1.
quandtatem , figuram, m otus, et proprietates alias nisi in respectu ad intelligendam , ut
27. “ O mnia aetema quae sunt non gencrabilia, sed m oventur m otu secundum translationem,
Com m en tator 2 D e caelo et mundo dicit. N o n ergo est forma quia form a non est in poten- habent materiam; sed non habent materiam generabilium sed materiam eorum que m ov­
tia ad suas postremas perfecdones; nec est materia quia materia est in potentia ad ac­ entur de ubi in u b i.” Averroes [Metaphysics, bk. 12, com m ent. 10], 1562—1574, 8:296V,

tum primum. E rgo est quasi m edium , ut sic, ratio subiecd et essentia coniuncta.” Ibid., col. 2—297r col. 1. For Aquinas’ s interpretation o f these passages, see W ippel, 1981a, 286—
(E -F ). 287.
250 THE CELESTIAL REGION 251
ON CELESTIAL M ATTER

II. Tw o rival theories in support o f the existence o f celestial Despite the seeming total fulfillment o f celestial matter by its single,
matter unique, and permanent form, one aspect o f potentiality remained. The
uniform, circular motion o f the planets and stars compelled Thomas, as it
i. Aquinas and Galileo: Celestial and terrestrial matter differ had Aristotle, to concede that a celestial body was in potentiality at least
with respect to place,33 even if not with respect to being. Although celestial
Arrayed against Averroes and all who denied the existence o f celestial matter and terrestrial matter shared a potentiality for place, they otherwise had
were Thomas Aquinas and Aegidius Romanus, who, despite their agree­ nothing in common. Unlike the prime matter o f the terrestrial region, which
ment that the heavens are composed o f matter and form ,28 disagreed rad­ was a pure potentiality that could receive and lose all possible forms existing
ically as to the nature o f that matter, Thomas assuming it radically different in that region, celestial matter was created with a single form so complete
from, Aegidius identifying it with, terrestrial matter. These tw o differing that it precluded the receipt o f any other possible forms. Thus was the
interpretations by Thomas and Aegidius, along with the opposing opinion
incorruptibility o f the heavens preserved.
o f Averroes, lay at the core o f almost all discussions o f the problem o f Whereas Thomas considered the problem o f celestial matter briefly, in
celestial matter to the end o f the seventeenth century. Because o f their many treatises, Galileo, w ho was in essential agreement with Thomas,
obvious importance, we shall examine them in some detail. treated the problem at great length in only one treatise. In Galileo’s questions
Although Thomas, and many who followed him, argued that the celestial on De caelo, at least four o f the questions are relevant to the existence and
region consisted o f a composite o f matter and form, the matter he had in nature o f celestial matter.34 As he subdivides the major opinions into a host
mind “ was o f another kind than that o f inferior [i.e., sublunar] bodies.” 29 o f confirming arguments and objections, Galileo is the quintessential scho­
In the heavens, matter was in potentiality with respect to a perfectly ac­ lastic. Embedded within this rather heavy format are Galileo’s ow n opin­
tualized form that fulfilled all the possibilities o f that matter, which therefore ions, accompanied by objections to those opinions that are systematically
lacked a potentiality for any other form s.30 Consequently, changes o f sub­ answered at the end o f each question. O f the great number o f extant dis­
stance, quality, or quantity could not occur in the heavens. The Sun, for cussions on the problem o f the possible existence and nature o f celestial
example, was incapable o f changing into anything else, nor did it come matter, Galileo’s must rank as one o f the most thorough. A ll but one o f
into being by the transformation o f anything else.31 B y contrast, “ the matter the opinions are described with scholastic fullness o f detail and subtlety.
o f the elements is in potentiality with respect to an incomplete form which And, as was characteristic o f sixteenth- and seventeenth-century scholastics
cannot limit [or fulfill] the potentiality o f the matter.” 32 - but not o f their medieval predecessors - Galileo cited a large number o f
authoritative sources; ancient, medieval, and Renaissance. Here we find
28. Galileo, D e caelo, qu. 5 (K), 1977. 1 11—112, cites numerous supporters o f this general
specific citations to the relevant works o f the supreme early authorities,
opinion, including “ all the Arabs, w ith the single exception o f A v e rro es,” specifically
mentioning A vem pace (ibn Bajja) and Avicenna; M oses M aim onides, Saint Bonaventure, Aristotle and Averroes, as well as to the works o f lesser early figures such
Thom as Aquinas “ and likewise all T h o m ists,” Aegidius Rom anus (Giles o f Rom e), A l- as Alexander o f Aphrodisias, Simplicius, John Philoponus, and Avicenna.
bertus M agnus, Alessandro Achillini, and Julius Caesar Scaliger.
Numerous medieval arguments are cited from eminent scholastic authors
29. Thom as Aquinas [De caelo, bk. 1, lec. 6, par. 63], 1952, 31, declares: “ quod materia
caelestis corporis est alia et alterius rationis a materia inferiorum corp o ru m .” Th is passage, such as Aquinas, John o f Jandun, Saint Bonaventure, Albertus Magnus,
and some thirty-six others (drawn from a variety o f T h o m a s ’s w orks) that reflect his Duns Scotus, Marsilius o f Inghen, Durandus de Sancto Porciano, and es­
view s on the relationship o f celestial and terrestrial matter, appear in Litt, 1963, ch. 3,
pecially Aegidius Romanus. From his contemporaries and predecessors o f
pp. 54-80; for the quotation, see p. 79; for similar statements from T h o m a s ’s Commentary
on the Metaphysics, and Quaestio disputata de anima, see Litt, ibid., 72, no. 22, and 77, no. the fifteenth and sixteenth centuries, Galileo invoked the opinions and works
31, respectively. For a fine, brief account o f T h o m a s’s view s, especially as they contrast o f Alessandro Achillini, John Capreolus, Cajetan (Thomas de Vio), Julius
with the opinions o f G od frey o f Fontaines, see W ippel, 1981a, 285-291. A lth ou gh he
makes no mention o f Thom as, Hervaeus N atalis, D e materia celi, qu. 3, 1513, 38V, col.
Caesar Scaliger, and Marsilio Ficino.
2, adopted the same position, dem onstrating it, h ow ever, not b y positive arguments but From this impressive parade o f authorities, w e should not infer that
b y the fairly com m on medieval practice o f refuting all o f its alleged rivals - in this case, Galileo had a deep familiarity with the vast literature on the problem ot
three other theories.
30. “ N am materia caelestium corporum est in potentia ad actum perfectum, idest ad formam
33. See Litt, 1963, 5 6 -57, especially note 3, w hich contains eight passages drawn from eight
quae com plet totam possibilitatem materiae, ut iam non remaneat potentia ad alias formas.
different treatises. O n page 56, Litt also cites four passages in A ristotle’s Metaphysics
Materia autem elementorum est potentia ad form am incom pletum , quae totam terminare
( 8 .1 .1024b.6; 8 .4 .1044b.7; 9-8.1050b.21; 1 2 .2 .1069b.2).
non potest materiae potentiam .” D e substantiis separatis c. 8, n. 82, as cited b y Litt, 1963,
34. These are questions no. 3 (I) (“ Are the heavens one o f the simple bodies or com posed o f
79 -
them ?” ), Galileo, D e caelo, 1977, 81-92; no. 4 (J) (“ A re the heavens incorruptible?” ),
31 . See Litt, 1963, 59; for T h o m as’s statement from his com m entary on B oethius’s D e trinitate,
ibid., 93-10 2; no. 5 (K) (“ A re the heavens com posed o f matter and form ?” ), ibid., 10 3 -
see page 63.
147; and no. 6 (L) (“ Are the heavens animated?” ), ibid., 148—58. T h e fifth question is the
32. For the Latin text, see note 30 in this chapter.
m ost relevant.
252 ON CELESTIAL MATTER 253
THE CELESTIAL REGION
yet another argument, w e are told that matter and form are the principles
celestial matter. As William Wallace has demonstrated, Galileo drew heav­
o f natural things: “ therefore, since the heavens are natural, they must be
ily, and perhaps wholly, upon a few published treatises and a larger number
o f unpublished lectures, or reportationes, that had been produced by the Jesuit composed o f matter and form ” (ibid., 117).
B y such arguments Galileo was convinced that matter o f some kind must
faculty o f the Collegio Romano between the 1570s and 1590s.35 Despite his
exist in the heavens. Like so many before him, he inquired next about the
heavy debt to the Jesuits o f the Collegio Romano, the selection o f the
nature o f that matter: Is it the same as, or different from, our terrestrial
arguments and the final organization o f the questions are probably Galileo’s.
variety? He concluded that “ the heavens are not composed o f matter o f the
If Galileo sincerely believed in the opinions that he adopted in these treatises,
same kind as the matter o f inferior bodies” (ibid., 132). T o defend this
one can only conclude that he was a deeply committed Aristotelian scholastic
position, Galileo found it necessary to attack in considerable detail the most
when he wrote them, a picture that stands somewhat at variance with the
important opposition thesis, namely Aegidius’s well-known arguments in
Galileo o f the other Juvenilia, the one w ho wrote the De motu and had clearly
defense o f the identity o f celestial and terrestrial matter (ibid., 13 2-13 9). In
broken with Aristotle on the problem o f motion in a vacuum.36
the process, Galileo turned Aristotle’s denial o f celestial matter to his own
In the question “ Are the heavens composed o f matter and form?” Galileo,
advantage by insisting that it was not to be taken categorically. It was only
as w e w ould expect in a scholastic treatise, considered both the negative
Aristotle’s w ay o f denying that “ there is in the heavens matter o f the same
and the affirmative positions. The champion for the negative side was A v ­
kind as the matter o f lower bodies” (ibid., 124; also 132). Galileo thus
erroes, who, as we%saw, had denied that the heavens were composed o f
convinced him self that he was in agreement with Aristotle when he assumed
matter and form .37 Following a lengthy description o f Averroes’ position
that celestial matter was something quite different from terrestrial matter.
(Galileo [De caelo, qu. 5 (K)], 1977, 1 0 3 - m ) , Galileo presented the case
If it existed in an incorruptible heaven, it had to be radically different -
for the affirmative side, which constituted the majority opinion during the
indeed, nothing less than incorruptible. Galileo, like all who adopted the
M iddle Ages and the Renaissance, when its supporters included such famous
same interpretation, was committed to a conception o f celestial matter that
authorities as Avicenna, Maimonides, Albertus Magnus, Aquinas, Aegi-
made it function in ways that were largely the opposite o f its terrestrial
dius, Bonaventure, Achillini, and Scaliger.38 Within the majority group,
counterpart. The tw o matters were as radically different as the heavens from
debate centered on whether celestial matter differed from terrestrial matter.39
O n the existence o f celestial matter, Galileo - “ following the common the earth.
The differences - and here Galileo relied on and agreed with Aegidius -
opinion o f the peripatetics” — aligned him self with those w ho believed that
derived from the causes o f corruption, which were always explained in
“ the heavens are composed o f matter and form, whatever the matter may
terms o f matter, form, and privation (ibid., 134—135). Corruption occurs
be” (ibid., 115). In support o f the claim for the existence o f celestial matter,
when matter that possesses a form seeks the contrary o f that form. The
Galileo presented numerous arguments, most o f which were variations on
privation o f that contrary form is thus identified as the cause o f the matter’s
a few basic medieval Aristotelian themes. For example, since our heaven is
corruption.4* The privation itself, however, “ arises from the fact that a form
conceived as a particular entity, which he terms “ this heaven,” and all
has a contrary” (ibid., 134). Unless that privation can be overcome, how ­
particular entities consist o f matter and form, the heaven must also be
ever, so that the matter can at some time possess that contrary form, it
composed o f matter and form. A brief syllogism drawn from the Metaphysics
would be perpetually frustrated and therefore opposed to the operations o f
constitutes a second argument: “ Sensible substances contain matter; but a
heavenly body is singular and sensible; therefore [it contains matter].” 40*In nature, which does nothing in vain.
If a form did not have a contrary, the matter that possessed that form
35. See Wallace, 1981, 281, 309. For a list o f the Jesuit authors on w h om Galileo seems to could not be in a state o f potentiality with respect to a contrary form.
have relied, see Galileo, D e caelo, 1977, 1 2 -2 1 . O f this group, Christopher C lavius is the
best know n.
Therefore that matter could not be deprived o f a contrary form, and pri­
36. Whether Galileo believed the opinions he presents in the Juvenilia is difficult to determine. vation w ould not, and could not, function as the cause o f corruption.
See m y review o f Wallace, 1981, in Science 214 (1981), 55-56 . Although such matter could not exist in the terrestrial region, Galileo
37. A s indicated earlier, Averroes occasionally spoke as i f matter and form existed in the
heavens, although not as a com posite.
declared that it did exist in the heavens. Celestial matter had only one form,
38. For these and other names, see Galileo, D e caelo, qu. 5 (K), 1977, x 1 1 - 1 1 2 . Galileo declares which lacked a contrary. W ithout a contrary, privation could play no role,
that all Arabian authors w ith the exception o f Averroes supported this opinion, as did and celestial matter could have “ no appetite for another form, for if it did,
all Th om ists. A n even larger list is furnished b y B artholom ew A m icus [D e caelo, tract.
4, qu. 1, dubit. 2, art. 2], 1626, 138, col. 2.
it would have an appetite to be deprived o f its ow n existence” (ibid., 135).
39. Galileo, ibid. 112. Galileo also identified a second major issue (ibid.) that turned on
whether the form that is associated w ith celestial matter is also an intelligence.
41. “ Hence the matter remains deprived, and is in potency to another form and therefore
40. Ibid., 116. T h e passage, where Aristotle says that “ sensible substances all have matter,”
has an appetite for it; then corruption results.” Ibid. 134.
appears in the Metaphysics 8.1. i042a.24-26.
254 THE CELESTIAL REGION ON CELESTIAL M ATTER 255
Celestial matter was thus incorruptible, because it had no inclination for thickness and transparency in the heavens except by [the assumption of]
any other form. N or did it have any inclination to destroy its only form, matter.” 48 From Aristotle’s declaration that “ everything that is perceptible
because that would indicate a desire for its ow n nonexistence, which is is in matter” and that our entire world is compounded o f matter,49 Aegidius
absurd (ibid., 141). concludes that our heavens are “ particular and sensible heavens” and nec­
Galileo did not consider his arguments for the existence o f an incorruptible essarily imply a form in matter.50 Convinced that matter must exist in the
celestial matter as demonstrative “ but only as highly probable; because with heavens, Aegidius, in the second question o f his De materia celi, determines
the single exception o f Averroes, it is that o f practically all the peripatetics whether the matter o f the celestial region is essentially the same as the matter
and because there is nothing that contradicts it, while there are many things o f the inferior, or sublunar, region and concludes that they are indeed
in its favor” (ibid., 124).42 identical. According to Aegidius, none o f the ancient doctors, saints, and
philosophers whose works have reached us was o f the opinion that matter
2. Aegidius and Ockham: Celestial and terrestrial matter are exists in the heavens and therefore none o f them believed that the heavens
identical are composed o f matter and form. Those among these venerable authorities
who did assume the existence o f matter in the heavens, however, did not
Although Aegidius Romanus agreed with Thomas Aquinas that the celestial think there was any difference between that matter and the matter here
substance was a composite o f matter and form, he differed from him by below, as “ some masters [magistri] and modern doctors [doctores])” assert.5'
arguing that heavenly matter was essentially the same as terrestrial matter. Bolstered by a conviction that ancient authorities who considered the
In De materia celi (On Celestial Matter), a treatise devoted solely to the prob­ problem allowed that if matter did exist in the heavens, it would be identical
lem o f the possible existence and nature o f celestial matter, tw o questions with its terrestrial counterpart, Aegidius insisted that the heavens are not
form the basis o f the work. In the first, Aegidius considers whether matter simple but are instead a composite entity made up o f matter and form, a
exists in the heavens43 and, on the assumption that it does, poses the second: judgm ent that followed from the admitted quantity, thickness, transpar­
Is it essentially the same as terrestrial matter?44 ency, and individuality o f the celestial region.52 But is the matter in this
T o show that the celestial substance is composed o f matter and form, compound o f matter and form a pure potentiality, or is it some kind o f
Aegidius directly opposes the opinion o f Averroes that the heavens are a
simple, incomposite, indeterminate, and indivisible entity - that is, a form pint o f water equals 10 pints o f air, a comparison that is possible only because air and
without matter. Aegidius stresses the evidence o f our senses, declaring that water share matter - not form - in virtue o f w hich one can say that water is “ thicker”
than air. T h e same kind o f relationship obtains betweeen stars and celestial orbs. Th e
we can observe that the heavens possess quantity and are therefore divisi­
ratio o f 10 to 1 between water and air was merely a hypothetical exam ple for Aristotle,
ble.45 But without matter, the heavens would be incapable o f receiving the although A egidius invokes it as a fact.
quantity necessary to make them divisible. Appealing again to sense,46 A e­ 48. “ E rgo non poterimus salvare spissum et dyaphanum in celo, sed solum ex materia.”
A egidius Rom anus, D e materia celi, 1 502b, 79V, col. 2.
gidius observes that one part o f the heavens is thicker than another, as
49. Aristotle, D e caelo, 1.9.278a.8 -1 6 [Guthrie], i960.
evidenced by the appearance o f stars, which are visible because they are 50. For a fuller discussion, see Grant, 1983, 16 5-16 7 .
thickened celestial matter in contrast to celestial orbs, which are rarefied to 51. “ D icen dum quod in hac questione sic procedemus quod primo ostendemus quod nulli
antiquorum doctorum nec philosophorum nec sanctorum de his qui pervenerunt ad nos
the point o f transparency.47 Aegidius concludes that “ w e cannot save the
secundum ea que vidim us fuerunt huius positionis: quod in celo esset materia et quod
corpus celi circumscripta intelligentia esset com positum ex duabus substantiis, ex materia,
42. A s we saw earlier, Averroes had Peripatetic followers in the M iddle A ge s. B y the late scilicet, et forma, et tamen materia ilia esset differens per essentiam ab tsta, sicut aliqui
sixteenth century, however, G alileo’s claim m ay have been correct. magistri et m odem i doctores posuerunt. Antiqui enim doctores vel negaverunt in celo
43 - “ Questio est utrum in celo sit materia vel sit celum corpus sim plex, ut posuit C o m m en ­ esse materiam, ut posuit C om m entator, vel si posuerunt ibi materiam dixerunt earn esse
tator. ” A egidius Romanus, D e materia celt, 1 502b, 78r, col. 2. For bibliographical refer­ eandem cum materia istorum inferiorum .” A egidius Romanus, D e materia celi, 1502b,
ences to the life and works o f Aegidius, see W ippel, 1981a, x i-x ii. 8ir, col. 1. A m o n g the saints w h o believed that celestial and terrestrial matter were
44 - “ Queritur secundo: dato quod in celo sit materia utrum ilia materia sit eadem per essentiam identical, A egidius mentions on ly Saint A ugustine, w hile he cites A vicenna as repre­
cum materia istorum inferiorum .” Aegidius Rom anus, ibid., 8ov, col. 1. sentative o f the philosophers (ibid., 8ir, cols. 1-2 ). In his discussion o f the question,
45. “ It is evid en t,” Aegidius insists, “ that the heavens have quantity, because to deny this W illiam o f O ckh am , asserted the opposite when, after declaring that “ there is matter in
men w ould have to hallucinate and deny the senses.” Ibid., 79V, col. 1. the heavens,” he justified this claim b y “ the statements o f the saints and doctors, w ho
46. Contrary to Aegidius, Francisco Suarez, Disputationes metaphysicae, disp. 13, sec. 10, 1886, say that, in the beginning, G o d created matter from w hich the celestial bodies and other
1:436, col. 2, in the late sixteenth century, denied that one could dem onstrate b y any things were form ed.” O ck h am [Sentences, bk. 2, qu. 18: “ U trum in caelo sit materia
visible effects whether the heavens were com posed o f form alone, as A verroes, w ould eiusdem rationis cum materia istorum inferiorum” ], 1981, 5:399.
have it, or o f matter and form, as Thom as argued (and, o f course, A egid ius, w h o is not 52. “ A d h u c celum non est corpus sim plex, ut posuit C om m entator, quod probabamus et ex
mentioned).
quantitate eius, et ex spissitudine, et ex dyaphanitate quas videm us in ipso et ex indivi-
47. T o reinforce his position, Aegidius appeals to the second b o o k o f A ristotle’s O n Generation duatione ipsius.” A egidius Romanus, ibid., 8 iv , col. 1. In this opinion. A egidius says
and Corruption (2.6.3333.21-24) where, according to A egidius, Aristotle declared that 1 that he does not differ from other theologians (nec in hoc discordamus ab aliis theologis).
256 THE CELESTIAL REGION ON CELESTIAL MATTER 257
actuality? Aegidius argues for the former, insisting that i f the matter were ing. ” 58But it does not share in any o f the properties o f those extremes. “ For it
actualized it could not form a single essence with the actualized form, is not properly a being, because it is not something in act; nor is it absolutely
“ because one thing is never formed essentially from tw o actualized nothing, because it is something in potentiality. ” 59 O r to put it another way,
things.” 53 Celestial matter must, then, be “ pure potentiality” {purapotentia), if matter “ became something that had less being than pure potentiality, it
and therefore identical with the matter o f inferior things.S4 would immediately become nothing, because it w ould be neither actual nor
Aegidius offered three arguments to support his claim that celestial and potential; but i f it had more o f being than pure potentiality, it would be nec­
terrestrial matter are essentially the same.55 The first relies on a “ principle essary that it become an actualized thing. ’,6° Thus matter can become neither
o f indifference, ” whereby Aegidius assumes that i f celestial and terrestrial o f these extremes without losing its unique status as pure potentiality. Be­
matter are both pure potentialities - as he believed — and if every form were cause the argument applies to both celestial and terrestrial matter, it follows
stripped from those tw o matters, they w ould not differ in any w ay, “ because that both are pure potentiality and must always remain so.
there can be no distinction in pure potentiality.” B y their “ indifference” For all these reasons, Aegidius concluded that not only does matter exist in
(per indifferentiam), or lack o f difference, the unity and identity o f the two the heavens but that as pure potentiality it is identical with the matter o f the
matters must be accepted.56 sublunar, or inferior, region. But, as Aegidius recognized, even i f this were
In the second defense o f the identity o f celestial and terrestrial matter, true, a major problem yet remained. If celestial and terrestrial matter are iden­
Aegidius attempts to compare celestial and terrestrial matter. He asks tical as pure potentiality capable o f receiving forms, w h y do generation and
whether matter that serves as the subject o f a higher form is not thereby corruption occur only in the sublunar region? In his response, Aegidius in­
more worthy, and therefore more actualized, than matter that is the subject vokes the three fundamental principles o f change: matter, form, and priva­
o f a less w orthy form. Since all w ould readily agree that celestial forms are tion, the last-mentioned serving as the contrary to a form. Generation and
nobler than terrestrial forms, w ould it not follow that celestial matter is corruption occur when all three are present. In the heavens, however, contrar­
more actualized than terrestrial matter, and therefore different from it? ies or privations o f a form are absent. Celestial forms lack any associated pri­
Aegidius denies the very basis o f the comparison by insisting that distinc­ vation (privatio admixta) and are, consequently, perpetual and incorruptible:
tions between matters that are pure potentiality could not arise from the
hierarchical status o f the forms which actualize those potentialities. If ce­ If, therefore, you wish to assign a cause as to why these [sublunar] things are
lestial and terrestrial matter are pure potentiality, no distinctions can be corruptible and not those [celestial] things, you should not assign this on the basis
assigned between them on the basis o f the greater or lesser nobility o f the of the diversity of matter, because the matter is essentially the same here and there.
forms they may support.57 But you should assign this [cause] based on the diversity of the form, because the
[sublunar] form has a contrary and [therefore] has an associated privation; but that
As his third defense o f the idea that all matter is pure potentiality, Aegidius
argues that as pure potentiality, matter is “ a mean between being and noth­ [celestial] form does not have a contrary, nor an associated privation. Thus these
[sublunar] things are corruptible, not those [celestial] things.61

53. “ N unquam ex duobus in actu sit unum per essentiam .” Ibid. The celestial and terrestrial regions operate in contrary ways. Here below, a
54. “ E o ergo ipso quod ponimus ibi puram potentiam , oportet quod ponam us ibi unam form with its associated privation is the cause o f corruption, even though mat-
materiam, et oportet quod ilia materia sit eiusdem rationis cum materia istorum infer­
iorum .” Ibid., col. 2. 58. “ Tertia via ad hoc idem sumitur ex eo quod materia est quid m edium inter ens et nihil.”
55. Galileo summarizes seven arguments in favor o f the identity o f celestial and sublunar A egid ius Rom anus, D e materia celi, 1502b, 82V, col. 1.
matter w hich he attributes to Aegidius (see G alileo, D e caelo, qu. 5 (K): “ A re the heavens 59. “ Ponim us in terminis materia istorum inferiorum est potentia pura et ideo est media inter
com posed o f matter and form ?” , 1977, 1 1 3 - 1 1 5 , pars. 42, 45, 4 7-4 9 , 52, 56). For G alileo’s ens et nihil. N o n enim est proprie ens quia non est quid in actu; nec est om nino nihil
rejection o f these seven arguments, seepages 138 -13 9 . O n pages 13 2 -13 6 , G alileo discusses quia est quid in potentia.” Ibid., col. 2.
other aspects o f A cgid iu s’s thoughts on this question. O n e m ay rightly infer that A e g i- 60. “ E t ut m agis dare ostendamus propositum dicamus quod potentia pura non potest per-
dius’s ideas were considered central in the debate on the relationship between celestial ficere in entitate nisi fiat ens in actu; nec potest in aliquo deficere in enritate nisi fiat nihil
and terrestrial matter. quia ex quo materia est pura. Si esset aliquid quod esset minus ens quam potentia pura
56. “ E t quia materia celi est potentia pura et materia istorum inferiorum est potentia pura, illud statim esset nihil quia nec esset actus neque potentia; sed si plus habeat de entitate
si absolverentur ab om ni forma, materia huius et ilia non haberent per quid differrent quam potentia pura oportet quod sit aliquid ens in actu.” Ibid.
quia in pura potentia non potest esse distinctio. C u m ergo dictum sit quod unitas materie 61. “ Si ergo vis assignare causam quare ilia sunt corruptibilia et non ista non assignes hoc ex
et identitas eius accipienda sit per indifferentiam eo ipso quod materia celi absoluta ab diversitate materie quia eadem est materia per essentiam hie et ibi. Sed assignes hoc ex
om ni forma non haberet per quid differret a materia istorum inferiorum sic absoluta, diversitate form e ut quia forma ista habet contrarium et habet privationem annexam; ilia
oportet quod sit una materia et eadem per essentiam et eiusdem rationis hie et ib i.” autem form a non habet contrarium nec privationem annexam. Ideo ista sunt corruptibilia,
Aegidius Romanus, D e materia celi, 1502b, 82r, col. 1. non ilia.” Ibid. 83r, col. 2. A s the source o f this opinion, Aegidius dtes A vicenna’s
57. For a fuller discussion and docum entation, see G rant, 1983, 16 8-16 9. Sufficientia, bo ok r.
258 THE C E L E S T I A L R E G IO N ON CELESTIAL M ATTER 259
ter, as pure potentiality, must receive that contrary. In the heavens, however, say the form o f fire or water, into celestial matter and thus subject the
the form cannot cause corruption because it lacks any contrary or privation. heavens to generation and corruption. In effect, the same substantial changes
Thus the purely potential matter, which is capable o f receiving another form, produced by natural agents in the terrestrial region could be caused in the
is never provided an opportunity to exercise that potentiality.62 celestial region by divine power.
Wielding his sharp, trusty razor, William o f Ockham also upheld the But Ockham even imagines a sitution where the celestial matter would
identity o f celestial and terrestrial matter,63 though for reasons quite different be acted upon naturally by a created agent. This might occur i f God intro­
from Aegidius’s. Ockham frankly admitted that the identity o f the two duces the form o f the element fire into celestial matter and if a quantity o f
matters was not demonstrable but quickly added that neither was any other water o f greater active power than fire was sufficiently near. Under these
opinion on this issue.64 H owever, by use o f his razor and the concept o f circumstances, the water could naturally corrupt the form o f fire and in­
G od’s absolute power, Ockham hoped to make his case the most persuasive. troduce the form o f water into celestial matter.67
Although the incorruptibility o f the heavens was routinely assumed, O ck­ Thus the mere possibility that God could — although he almost certainly
ham insisted that celestial incorruptibility was not absolute, because if it would not - effect the same substantial changes in the heavens that are caused
pleased God, he could corrupt and destroy the heavens. B y celestial incor­ by natural agents in sublunar things, led Ockham to conclude that “ the mat­
ruptibility, then, we must mean that the heavens are not corruptible by any ter in the heavens is the same kind as in inferior things.” For w hy should we
created agent. Thus the difference between terrestrial matter and celestial assume tw o different kinds o f matter when one will do? “ A plurality is never
matter reduces to this: in the former, God and/or some created agent has to be posited without necessity,” Ockham insists. T w o varieties o f matter
the power to corrupt one form and generate another, whereas in the latter ought not to be introduced, “ because all things which can be saved by the di­
only God can corrupt or destroy the form. “ Therefore,” Ockham argues, versity o f the nature o f matter can be saved equally well, or better, by the un­
“ whether the matter is o f the same kind or o f a different kind does not ity o f the nature [of matter].” 68 Because only one kind o f matter exists in
affect its corruptibility or incorruptibility.” 65 And even though the matter heaven and earth, and the matter in the latter is subject to generation and cor­
in each region is corruptible under different circumstances, the potentiality ruption, w e may infer that for Ockham the heavens were also subject to gen­
for corruption is always there. The “ difference between the matter here and eration and corruption, even though no potentially corruptive forms might
there,” Ockham explains, is that “ the matter here is in potentiality to other ever appear there to cause an actual generation and corruption.
forms that can be produced by a natural, created agent and [also] to some
which can only be created by God alone, as, for example, the intellective
form. But the matter o f the heaven is in potentiality to many forms, none III. Celestial matter in the late sixteenth and the seventeenth
o f which can be produced or induced in that matter by a natural agent but century
only by G o d ,”66 who, if he wished, could introduce any terrestrial form,
j. The focus o f the debate: Thomas or Aegidius?

62. “ Dicem us itaque quod econtrario contingit de corruptione in istis inferioribus et de cor- The impact o f these tw o rival thirteenth-century theories continued strong
ruptione in supercelestibus. N a m hec form a si est causa corruptionis hoc est propter
materiam; ibi autem est causa incorruptibilitatis propter formam. T o tu m tamen hoc
throughout the Middle Ages69 and into the sixteenth and seventeenth cen-
contingit quia in istis inferioribus est adm ixta privatio; in supercelestibus vero est carentia
privationis. Ideo ilia sunt perpetua; hec autem corruptibilia. ” Ibid. multas quarum nulla per agens naturale potest produci nec induci in ilia materia, sed
63. In his com m entary on the Sentences, bk. 2, qu. 18: “ U tru m in celo sit materia eiusdem solum a D eo possunt ista fieri.” Ibid., 403.
rationis cum materia istorum inferiorium. ” See O ckh am , Sentences, bk. 2, qu. 18, 1981, 67. “ N a m materia caeli, ex quo est eiusdem rationis cum materia hie, est in potentia non
5:395-409- tantum ad illas formas quae solum possunt causari a D eo et non .ab agente creato, sed
64. “ Secundo dico quod in caelestibus et in istis inferioribus est materia eiusdem rationis etiam ad formas quae possunt produci ab agente creato, puta ad formam ignis, aeris, etc.
om nino, licet haec pars non possit demonstrari sicut nec alia.” Ibid., 400. Such declarations Posito igitur quod Deus in materiam caeli induceret formam ignis, sicut est possibile
were not uncom m on. For similar statements in the sixteenth century, see John M ajor quia non includit contradictionem, si aqua tunc esset approximata et esset maioris virtutis
[SenrcMces, bk. 2, dist. 12, qu. 3], 1519b, 65r, col. 2, and 66v, col. 1, and Suarez, Dis- in agendo quam ignis, corrumperet formam* ignis et induceret formam aquae in materia
putationes metaphysicae, disp. 13, sec. 10, 1866, 1:436, col. 2; for the seventeenth century, quae prim o erat sub forma caeli. Et ideo materia caeli simpliciter est in potentia passiva
see Riccioli, Almagestum novum, pars post., bk. 9, sec. 1, ch. 5, qu. 2, 1651, 234, col. 2, ad multas formas quas potest agens creatum producere.” Ibid., 403-404.
where Riccioli also attributes this attitude to the Conim bricenses, Hurtado de M endoza, 68. “ Sic igitur videtur mihi quod in caelo sit materia eiusdem rationis cum istis inferioribus.
and Roderigo de Arriaga. E t hoc, quia pluralitas nunquam est ponenda sine necessitate, sicut saepe dictum est. N un c
65. “ Ergo quod materia sit eiusdem rationis vel alterius nihil facit ad corruptibilitatem vel autem non apparet necessitas ponendi materiam alterius rationis hie et ibi quia omnia
incorruptibilitatem.” O ck h am , Sentences, bk. 2, qu. 18, 1981, 5:401-402. quae possunt salvari per diversitatem materiae secundum rationem possunt aeque bene
66. “ Et in hoc est differentia inter materiam hie et ibi quod materia hie est in potentia ad alias vel melius salvari secundum identitatem rationis.” Ibid., 404.
formas quae possunt produci per agens naturale creatum et ad alias quae non possunt 69. N ic o le O resm e presents an anomalous situation. A lth ough he seems to have accepted
creari nisi a solo Deo, puta formae intellectivae. Sed materia celi est in potentia ad formas celestial matter as a com posite o f matter and form (Oresme, D e celo, bk. 1, qu. 11, 1965,
260 THE CELESTIAL REGION ON CELESTIAL MATTER 261

tunes. M ost early modem scholastic authors rejected the opinion o f A v­ exist in the heavens. Suarez argued (Disputationes metaphysicae, disp. 13, sec.
erroes and his followers and were agreed that the celestial region consisted 10, 1866, 1:437, col. 1) that all mobile physical bodies are natural beings
o f an ether that was composed o f matter and form .70But within this majority and fall under the domain o f philosophy. He observes that in De caelo, book
group there was the same split between those who assumed the identity o f 1, chapter 2, Aristotle himself classifies the celestial bodies in the category
celestial and terrestrial matter and those who assumed a radical difference. o f natural beings. But in the first tw o chapters o f the second book o f his
The overall number o f arguments for each opinion seems to have in­ Physics, Aristotle explains that nature is nothing but matter and form. In­
creased. M any o f the traditional medieval arguments for the leading opin­ deed, w e do not even know another kind o f physical nature. Suarez, there­
ions were repeated and often elaborated. However, some elements were fore, concludes that celestial bodies, as natural, physical beings, must be
introduced that either played little role in the Middle Ages or were wholly composed o f matter and form. A common scriptural argument invoked by
new. In this regard, Tycho Brahe’s rejection o f hard, solid spheres and the Bartholomaeus Mastrius and Bonaventura Bellutus ([De caelo, disp. 2, qu.
gradual shift to the concept o f fluid heavens played a role. 2, art. 2], 1727, 3:492, col. 2, par. 57) emphasized that on the first day o f
The case for a matterless heavens in the traditional sense ascribed to creation God created all the matter from which the whole world, including
Averroes and his followers seems to have steadily lost support.71 B y the both terrestrial and celestial components, was made; therefore, matter must
sixteenth and seventeenth centuries, additional arguments in favor o f a exist in the heavens.
heavens filled with a composite o f matter and form were available. It was The interpretation that the heavens consist o f matter and form easily
almost taken for granted that accidents such as quantity, along with rarity triumphed over the matterless concept o f Averroes and his followers. The
and density, that were common to terrestrial bodies also existed in the real struggle concerned the nature o f that matter: Was it identical with
heavens.72 Such properties were taken as evidence that matter must also terrestrial matter, as Aegidius and his followers maintained, or was it rad­
ically different, as Thomas Aquinas and his supporters would have it? Em­
162-164), he rqects A egid ius’s identification o f celestial and terrestrial matter (164-166);
inent scholastics o f the sixteenth and seventeenth centuries could be found
rqects the idea that a com pletely perfect form is associated w ith celestial matter and
negates the need o f the latter for other form s (166-168); and rqects T h om as Aquinas’s on both sides o f this issue. Thus Francisco Suarez (Disputationes metaphysicae,
theory that celestial matter is o f another nature than terrestrial m atter (168). B y both disp. 13, sec. 11, 1866, 1:440, col. 2), Raphael Aversa ([De caelo, qu. 33,
denying and affirming that celestial and terrestrial matter are identical, O resm e seems
sec. 6], 1627, 105, col. 2-109, col. 1), and Bartholomew Amicus ([De caelo,
caught up in a contradiction.
70. Riccioli, Almagestum novum, pars post., bk. 9, sec. 1, ch. 5, 1651, 232-233, distinguishes tract. 4, qu. 3, dubit. 1, art. 3], 1626, 146, col. 2) followed Thomas and
three com m on opinions: that the heavens are a simple b o d y not com posed o f matter and differentiated between celestial and terrestrial matter, whereas Mastrius and
form; that the heavens are com posed o f m atter and form; and a third opinion “ that each
Bellutus ([De caelo, disp. 2, qu. 2, art. 2], 1727, 3:492, col. 2, par. 57), as
o f the first [tw o opinions] is probable” and attributes the third opinion to Duns Scotus
[Sentences, bk. 2, dist. 14, qu. 1], Raphael A versa [D e caelo, qu. 33, sec. 5], 1627, 101, well as Giovanni Baptista Riccioli ([Almagestum novum, pars post., bk. 9,
cols, x—2, and Mastrius and Bellutus [D e coelo, disp. 2, qu. 2, art. 1], 1727, 490, col. 2 - sec. 1, ch. 5, qu. 2], 1651, 235, col. 1), and Melchior Comaeus ([De caelo,
492, col. 2. Riccioli mentions that the third opinion derives from the fact that one may
tract. 4, disp. 2, qu. 1], 1657, 488-489) assumed that they were identical.73
argue from the authority o f Aristotle that the heavens are not a com posite o f matter and
form, whereas on the authority o f theologians one m ay argue that they are. Scotus does In this debate, the theory that preserved the popular Thom istic distinction
this in the very question cited b y Riccioli, except that Scotus substitutes philosophers for between celestial and terrestrial matter not only perpetuated the traditional
Aristotle. John Major, Sentences, bk. 2, dist. 12, qu. 3, 1519b, 65r, col. 2, was not unusual
division between the celestial and sublunar realms, but also preserved the
in declaring as the first o f five conclusions that “ although in the natural light it cannot
be effectively proved that the heavens do not have matter and form , nevertheless, in the principle o f celestial incorruptibility. The contributions o f T ycho Brahe and
natural light the opposite [that the heavens do have matter and form] is more apparent.” Galileo during the late sixteenth and the first quarter o f the seventeenth
71. A good barometer o f its waning popularity is the list o f supporters assigned b y Riccioli
century seriously challenged both o f these ideas. For scholastics who, where
to the tw o major opinions. His list for A verro es’ opinion contains perhaps one seventeenth
century figure ([Philip?] Faber), whereas he mentions approxim ately six seventeenth- possible and feasible, were desirous o f adapting and adjusting scholastic
century supporters for the rival theory that the heavens consist o f matter and form. See
Riccioli, Almagestum novum, pars post., bk. 9, sec. 1, ch. 5, 1651, 232, col. 2 -23 3 , c°l-
1. T o Faber, w e m ay also add T h om as C o m p to n -C a rleto n [D e coelo, disp. 3, sec. 3], ch. 2, qu. 4], 1598, 40; Suarez, Disputationes metaphysicae, disp. 13, sec. 10, 1886, 1:437,
1649, 406, col. 2, w ho after declaring, “ it is m ore probable that the em pyrean heaven col. 1; and M ajor, Sentences, bk. 2, dist. 12, qu. 3, 1519b, 65V, col. 2. A egidius h im self
does not consist o f matter and form but is a simple b o d y ” adds, in the next paragraph, had emphasized the existence o f accidents in the heavens that were norm ally associated
“ it seems more probable to me that all the heavens are not com posed o f matter and form w ith matter and that were therefore go o d indicators that matter existed in the heavens.
but are similarly simple bodies. ” A lth o u gh C o m p to n -C a rle to n did not believe that it was T h e kind o f “ rarity and density” that m ost scholastics had in m ind was o f the static (i.e.,
impossible that matter could exist in the heavens, he thought it implausible, because permanent) kind rather than the dynam ic (or variable) type (see C h . 10, Sec. II. i.e). Their
whereas matter and form in terrestrial bodies w ere productive o f continuous, obvious sense o f quantity, h ow ever, must have been o f the terrestrial kind.
change, no such changes were visible in the heavens. H e concludes, “ w e have no basis 73. Riccioli expresses his choice this w ay: “ A lth ough w e cannot k n o w dem onstratively and
for asserting that in fact there is matter in them [i.e. the heavens].” Ibid., 407, col. 1. evidentally w hat the visible substance and nature o f the heaven is, it is nevertheless more
72. See Aversa, D e caelo, qu. 33, sec. 5, 1627, 104, col. 2; Conim bricenses [D e coelo, bk. 1, probable that it consists o f matter that is identical w ith elementary m atter.”
262 THE CELESTIAL REGION ON CELESTIAL M ATTER 263

cosmology to the latest scientific knowledge exemplified in the relevant unknown until the sixteenth century, as were the details o f his interpretation,
w ork o f Tycho and Galileo, the theory that identified celestial and terrestrial which included an attack on Aristotle’s ether, or fifth element, and the
matter was the only hope o f achieving a degree o f accommodation with attribution o f a fiery nature to the Sun and stars.76*But, as we saw at the
the emerging new cosmology. beginning o f this chapter, the advent o f Aristotle’s natural philosophy ren­
For it was only by the assumption o f a single cosmic matter that the rigid dered ideas about terrestrial elements in the heavens obsolete. Indeed, ce­
division between the celestial and terrestrial regions could be destroved, lestial corruptibility and the assignation o f terrestrial elements to the heavens
along with its associated idea that the different, nobler, and more perfect were discussed only for the purpose o f refutation, as Hervaeus Natalis (ca.
celestial region was incorruptible. A nd yet in his version o f the identitv 1260—1323) did in his De materia cell."
theory, Aegidius Romanus retained the rigid division o f the celestial and Alm ost four centuries after Hervaeus, during the second halt ot the sev­
terrestrial regions and the incorruptibility o f the former. He could do this enteenth century, at least four scholastic authors - Giovanni Baptista Riccioli
on the basis o f the standard assumption that celestial forms had no contraries, (1598-1671), Melchior Cornaeus (1598-1665), George de Rhodes ( i 597 ~
whereas terrestrial forms did. Thus, although the matter was identical, the 1661), and Franciscus Bonae Spei (1617-1677), all Jesuits except Bonae Spei,
lack o f contrary forms in the heavens prevented the kind o f change that who was a Carmelite - concluded that the heavens were corruptible. Ot
occurred in the terrestrial region.74 A dramatic conceptual change occurred the four, Riccioli is by far the most significant and best known. In his New
only when part or all o f celestial and terrestrial matter was made identical Almagest (Almagestum novum) ot 1651 (bk. 9, sec. 1), Riccioli considers
in more substantial ways than in Aegidius s widely held theory, where the “ The Creation and Nature o f Celestial Bodies,’’ within which context - in
two matters were considered identical only as pure potentialities but other­ chapters 5 and 6 - he discusses the problems relevant to our subject. The
wise radically different, the one incorruptible, the other corruptible. Sig­ fifth chapter79 is devoted to the nature o f celestial matter - whether it is a
nificant changes followed upon the newly emerging cosm ology, which was simple or composite body and whether it is the same as or different trom
based on the consequences o f the Copernican theory and the particular elemental matter. The decisions on these questions are relevant to the sixth
celestial discoveries o f Tycho and Galileo, who revealed a dramatically chapter,80 in which Riccioli specifically asks “ Whether the heavens are ge­
different kind o f heavens from the heavens as described in traditional me­ n e ra te and corruptible” (An caelum sit generabile et corruptibile).
dieval cosmology. The heavens that Tycho and Galileo described could After following the usual scholastic procedure and presenting the argu­
bring forth comets, new stars, and variable sunspots. In this altered at­ ments pro and con for the various relevant questions, Riccioli concludes
mosphere ol cosmological speculation, some scholastics assigned to celestial not only that the heavens are composed o f matter and form - a popu­
matter the same fundamental properties as one or more o f the terrestrial lar opinion (as we saw) in the sixteenth and seventeenth centuries - but
elements. Despite centuries ot assumed celestial superiority, the great change that celestial matter is probably the same as the matter ot the sublunar re­
involved a terrestrialization o f celestial matter, whereby it came to be con­ gion,81 a thesis, as we noted, defended in the Middle Ages by Aegidius and
ceived as corruptible. Ockham.
From his assumption o f the essential identity o f celestial and terrestrial
matter, Riccioli drew a radically different consequence: ot the tour elements
2. Scholastic repudiation o f incorruptibility: the corruptibility o f
celestial matter burskv, 1962, 154—166; tor additional information, including a statement on celestial
corruptibility bvjohn Damascene, see Chapter 10, note 10.
The idea o f a corruptible heavens was o f ancient vintage and was well known 76. Samburskv, 1902, 158.
during the Middle Ages, when it was most prominently associated with 77. For a summary o f his critique, see Grant, 19S3, 163-164. O f the four elements. Hervaeus
thinks that only tire would be a plausible candidate as a celestial element. But it too tails
the name ofjo h n Philoponus, the sixth-century Christian Neoplatonist and to qualify because its naturally active qualities would consume the surrounding matter,
critic ot Aristotle. 5 Unfortunately, the relevant works ot Philoponus were nor could it account for the multiplicity o f effects that the heavens were assumed to cause,
such as coldness, wetness, and putrefaction.
74. For the details ot Aegidius’s theory, see Section II.; o f this chapter. 78., Book 9 is in the second part (pars secunda or pars posterior) o f the first volume, the only
75. In his commentary on De caelo, Thomas Aquinas, De caelo. bk. 1, lec. 6, 1952, 29, remarks / one o f the proposed three volumes to be published.
that Philoponus was one ot those who assumed that the heavens were, by their nature, 79. Riccioli, Almagestum novum, pars post., bk. 9, sec. 1, ch. 5, qus. 1-3, 1651, 232-236.
subject to generation and corruption. Thomas does not mention the source ot his mtor- 50. Ibid., ch. 6, 257-238.
mation, which was probably Simplicius s Commentary on De caelo !in the translation trom 51. Riccioli presents his opinion in the following conclusion: “ Licet non possit a nobis de­
Greek to Latin by William ot Moerbeke in 1271). For Simplicius's remarks, see Simplicius monstrative atque evidenter sciri, quaenam sit caeli visibilis substantia et natura, proba-
[De celo, bk. 1, comment. 20], 1540* i i v , col. 2 and I2r, col. 1. For a discussion ot bilius tamen est lllud constare ex materia eiusdem rationis cum elementari.” Ibid., ch. 5,
Philoponus's opinion in favor o f the identity o f celestial and terrestrial matter, see Sam- qu. 2, 235, col. 1.
264 THE CELESTIAL REGION ON CELESTIAL MATTER 265
(earth, fire, air, and water), the heavens must be composed o f one or more tion.” 88 But Riccioli informs us later that it was not only ideas from the
o f three o f the elements (fire, air, and water).82 Thus where Aegidius made Church Fathers and Scripture that led him to accept celestial corruptibility
prime matter the basis o f the identity o f celestial and terrestrial matter but also “ the arguments derived from experience concerning spots and
Riccioli made elemental matter the basis - that is, matter that had already torches near the solar disk that were discovered by the telescope and from
been actualized beyond the level o f prime matter. It was thus essential that certain comets that have come into being and passed away above the Moon.
Riccioli identify the elements o f which the heavens were composed. He These changes are more naturally explained by generation and corruption
concludes, for example: “ It is more probable that the heaven in which the than by other more violent means or by nonviolent miracles.” 89
fixed stars are is watery; the heaven in which the planets are is fiery.” 83 Perhaps because he was aware o f how radically he had departed from
Riccioli readily admitted that no genuine evidence or precise arguments traditional Aristotelian cosmology, Riccioli sought to salvage a remnant o f
could be offered in support o f the claim that the heaven o f the fixed stars celestial incorruptibility. Although, by their elemental nature, the heavens
was a congealed, watery solid and the heaven o f the planets a fiery fluid.*4 are intrinsically corruptible, they are not corruptible by any naturally created
Patristic authorities were, however, at hand. Some Fathers had held that external agent. Thus, for Riccioli, the celestial region was “ accidentally
the heaven consisted o f elemental water and others that it was composed incorruptible” (per accidens esse incorruptibile), because no natural, external
o f elemental fire.85 It therefore seemed a good compromise to identify the agent could corrupt it. This immunity from corruption by natural agents
sphere o f the fixed stars as the solid and watery sphere (both because the was perhaps a consequence o f the heavens’ great distance from the terrestrial
stars themselves remained fixed and unchanging and seemed to enclose the region, which was external to them, or perhaps attributable to the great
world and because the term firmamentum was used to describe the starry mass o f the celestial region, or possibly the result o f the distinctive nature
sphere) and to interpret the heaven through which the planets moved as a o f the primary qualities that God had placed in the heavens.90
fiery fluid, since the paths o f the planets varied.86 W hatever the reason, Riccioli’s concession to incorruptibility was o f little
Riccioli’s assumption o f a fluid planetary heaven was not o f itself a suf­ consequence, as is evident when he likens celestial incorruptibility to that
ficient indication o f a belief in celestial corruptibility,87 but his assertion that incorruptibility which applies to the whole earth and to the totality o f air,
these heavens consisted o f tw o terrestrial elements was. In his chapter on each o f which was really incorruptible as a totality even though its parts
the corruptibility or incorruptibility o f the celestial region, which follows suffered continual change. Despite their overall incorruptibility, parts o f the
immediately after the chapter that identifies celestial and terrestrial matter, heavens were nevertheless capable o f suffering corruption. In this the ce­
Riccioli declares the corruptibility o f the celestial region. O n the basis o f lestial region was just like the earth or air: it suffered change in its parts,
his assumption that the heaven o f the fixed stars is most probably watery but the whole endured unchanged.
and that the heaven o f the planets is fiery, he infers “ that from its very O n ly w ith regard to the empyrean sphere did Riccioli accept the tradi­
internal nature, the heavens have the capacity for generation and corrup­ tional opinion o f incorruptibility. The empyrean sphere was not, however,
a visible sphere, although it was required for the perfection o f the universe
82. B y om itting any discussion o f earth as a possible com ponent o f celestial matter, Riccioli
indicates his rejection o f it. and for the incorruptibility and eternal well-being o f our bodies.9'
83. “ Probabilius est caelum in quo sunt stellae fixae aqueum; caelum autem in quo sunt A m ong the other three seventeenth-century scholastic authors who re­
planetae igneum esse.” Riccioli, Almagestum novum, pars post., bk. 9, sec. 1, ch. 5, qu.
jected celestial incorruptibility, Melchior Cornaeus did it in a single para­
3, 1651, 236, col. 1.
84. In chapter 7 (ibid., 238-244), Riccioli considers the question “ W hether the heavens are graph.92 Because Cornaeus believed in the identity o f celestial and sublunar
solid or whether, indeed, some or all are fluid” (An caeli solidi sint, an vero fluidi omnes matter, and because the latter is a principle o f generation and corruption,
vel aliqui). A t the end o f the question, in a “ unica conclusio,” Riccioli declares that (ibid.,
he inferred that the heavens were corruptible.93 Partial corruption in the
244, col. 1) “ although it is scarcely evident mathem atically or physically, it is much more
probable that the heaven o f the fixed stars is solid, that o f the planets fluid.” In Chapter
88. “ Sequitur caelos hosce esse ab intrinseco et natura sua generationis et corruptionis ca-
14, w e shall consider medieval and early m od em scholastic ideas about the hardness or
paces.” Riccioli, Almagestum novum, pars post., bk. 9, sec. 1, ch. 6, 1651, 238, col. 1.
fluidity o f the celestial ether.
89. Ibid., 237, col. 2.
85. Ibid., ch. 5, qu. 1, 233, col. 2.
90. “ Q u ia tamen sive propter distantiam ipsorum, sive ob ingentem m olem , sive ob tem -
86. See ibid., qu. 3, 236, cols. 1 -2 .
peram entum insigne qualitatum secundarum cum primis quod Deus caelo indidit, non
87. A number o f scholastics had argued that the fluidity or solidity o f the heavens had no
datur agens ullum naturale creatum quod possit caelos transmutare substantialiter; idcirco
relevance to their corruptibility or incorruptibility. For example, see A m icus, D e caelo,
dixi per accidens esse incorruptibiles.” Ibid., 238, col. 1.
tract. 5, qu. 5, art. 1, 1626, 270. col. 2; Johannes Poncius, D e coelo, disp. 22, qu. 5, 1672,
91. Ibid.
620, col. 1; and Franciscus de O v ie d o [D e caelo, contro. 1, punc. 2], 1640, 1:462, par. 2.
92. Cornaeus, D e coelo, disp. 2, qu. 1, dub. 3, 1657, 489.
O vied o , indeed, believed that the heavens w ere both fluid and incorruptible. Ibid., 464.
93. In the next section, Cornaeus rejects the existence o f a celestial ether, or fifth element,
col. 1, par. 17.
and suggests that fire is the m ost probable matter o f the heavens. Ibid., dub. 4, 490-491.
266 ON CELESTIAL M ATTER 267
THE CEL EST IAL REGION

heavens was also evident from the many new stars that had been reported substantial generation and corruption could and did occur in the celestial
from as far back as 125 b . c ., including those o f 1572, r6oo, and 1604. region. In answering the charge that Aristotle had declared the heavens to
Corruptibility was also implied by scriptural predictions o f a Judgment Day be immutable and incorruptible, Cornaeus even declared that
in which the heavens will be destroyed. And finally Cornaeus argued that
the Sun, the most beautiful part ot the heavens, was corrupted almost daily if Aristotle were alive today and could see the alterations and conflagrations that
by fires. we now perceive in the Sun, he would, without doubt, change his opinion and join
George de Rhodes went beyond Riccioli and argued for the fluidity o f us. Surely the same could be said about the planets, of which the Philosopher knew
the entire heavens, including the sphere o f the fixed stars.94 Like Riccioli, no more than seven. But in our time, through the works of the telescope, which
however, he believed that although parts o f the heavens were corruptible, was lacking to him, we know tor an absolute certainty that there are more.'"’
the heavens, taken as a whole, were incorruptible.9'1 And also like Riccioli,
he judged the empyrean sphere to be absolutely incorruptible, whereas all Even some traditionalists like Aversa were prepared to break with Aristotle
other planets and spheres were corruptible. De Rhodes mentions Tycho, and allow that new stars and sunspots are celestial, rather than terrestrial,
who, he explains, showed that the new star o f 1572 was truly in the heavens. phenomena.
De Rhodes concludes that new stars could appear only in the celestial region, But w hy did scholastic Aristotelians yield on this seemingly important
because they are newly generated there. Thus generation as well as cor­ element in Aristotle’s cosmology? O n this, we can only speculate. Although
ruption can occur in the heavens.96 many scholastics denied a celestial location to the new phenomena, others
Following a series o f appeals to Scripture in favor o f celestial corrupti­ must have realized, as did Aversa, that astronomical data from the most
bility, Franciscus Bonae Spei grounds his belief in celestial corruptibility on respected astronomers o f the day could not be ignored indefinitely. Thus
“ the various generations and corruptions in the heavens revealed by the the first breakthrough — to concede the celestial location o f the new phe­
most certain observations o f the mathematicians’’ — that is, astronomers. nomena — was probably made rather painlessly, because it was still teasible,
Indeed, Bonae Spei insists that it is safer to accept the observations made and even easy, to insist that such phenomena represented only accidental
with the telescope and other instruments than to follow the philosophers, rather than substantial changes.
who, “ because o f the very great distance [of the sky] and the weakness o f The eventual transition to the concept o f celestial corruptibility was prob­
their eyes, are easily deceived.” 97* ably aided in no small measure by a widespread belief in the sixteenth and
In the seventeenth century, scholastic opinions about celestial incorrup­ seventeenth centuries that Plato, Scripture, and many Church Fathers were
tibility changed rather dramatically from what they had been during the agreed that the heavens were composed o f one or more terrestrial elements
period between the Middle Ages and the end o f the sixteenth century. Even and that the heavens were therefore capable o f substantial change.99 Indeed,
if the majority o f seventeenth-century scholastics retained the traditional as we saw, the eighth sphere o f the fixed stars was often identified as the
opinion — and this is by no means certain — those like Riccioli, Cornaeus, frozen or crystalline form o f the scriptural waters above the firmament.
de Rhodes, and Bonae Spei were prepared to abandon it and concede that Other Church Fathers had followed the Platonic idea that the heavens were
made o f the fourth and purest element, fire. Although such ideas were
94. De Rhodes [De coeio. disp. 2. qu. 1, sec. a], 1671, 278, col. 2-280. col. 1. De Rhodes known during the late Middle Ages, Aristotle’s conception o f a fifth in­
specifically refutes the explanations that new stars are not "new ” but have been in the
heavens all the time but are seen only when they become sufficiently dense fa view corruptible element, or ether, had replaced Platonic and patristic interpre­
attributed to Vallesius) and that new stars are produced by an accidental generation of tations.
opacity (a theory he rightly attributes to A versa; see ibid.. 279. col. 1). Since de Rhodes In addition to the works containing ideas about the corruptibility ot the
died in 1661 and his work was first published in 1671, the actual date o f composition is
unknown. heavens that had been available since the Middle Ages, others became avail­
95. “ Coelum licet corruptible sit, nunquam tamen posse corrumpi to turn; element.! enim able in the sixteenth century; for example, the works o f Plato, Philoponus,
tamersi sunt corruptibilia, semper tamen integra perstant sine ulla linminutione. Contin- and Saint Basil. In the developing anti-Aristotelian climate o f the sixteenth
gunt ergo in partibus coeli saepe mutationes.” Ibid., 279. col. 2. For what it is worth. /
de Rhodes makes no mention o f Riccioli in this section.
96. “ Dico primo celum empvreum omnino incorruptible; coeios autem sidereos esse cor- 98. “ Si Aristoteles hodie viveret et quas modo nos in sole alterationes et conflagrationes
ruptibiles.” Ibid., 278, col. 2. deprehendimus. videret absque dubio mutata sententia nobiscum faceret. Idem sane est
97. “ Probatur secundo ratione variae contingerunt generationes et corruptiones in coelis, ut de planetis quos Philosophus septenis plures non agnosat. At nos hoc tempore opera
constat ex certissima mathematicorum observatione, quos utpote per tot saecula obser- telescopii (quo ille caruit) plures omnino esse certo scimus.” Cornaeus [De coeio. disp. 2,
vatiombus et tubis opticis ahisque instrumentis, certissime collimantes tutius est sequi qu. 3, dub. 3], 1657, 303.
quam philosophos qui, propter longissimam distantiam et visus oculorum per se debi- 99. Riccioli, Alma^estum novum, pars, post., bk. 9, sec. 1, ch. 0, 1651, 237, col. 1-238. col.
litatem, facile decipiuntur.” Bonae Spei [De coeio, comment. 3, disp. 3, dub. 4]. 1652. 1, discusses all three and provides a lengthy list o f passages from the Bible and the Church
to, col. 2. Fathers.
268 THE CELESTIAL REGION
ON CELESTIAL MATTER 269
century, celestial corruptibility was an opinion that became more difficult
an anticipation o f the final rejection in the seventeenth century o f the tra­
to ignore than during the Middle Ages. The dramatic celestial discoveries
ditional distinction between the heavens and the earth. Such a temptation
o f the late sixteenth and the early seventeenth century provided the scientific
should be resisted. Aegidius’s identification o f celestial and terrestrial matter
basis for abandoning incorruptibility. Scholastics w ho found the celestial
as pure potentialities was made in the context o f an Aristotelian universe
discoveries o f Tycho and Galileo compelling could justify support for ce
in which the celestial region was assumed incorruptible and thus utterly
lestial corruptibility by direct appeal to Plato and, more significantly to
superior to a continually changing terrestrial matter.
the Church Fathers. O r they may have found the astronomical arguments
It was not Aegidius but John Philoponus who anticipated the seventeenth-
compelling only because o f the corroborating statements o f the Church
Fathers. century concept o f a universal matter everywhere subject to change. During
the Middle Ages only the idea was known, but not the works o f Philoponus
Although I shall discuss in Chapter 14 the transition from the concept o f
in which it was developed and justified. The dominance o f Aristotle’s phys­
solid heavens to that o f fluid heavens, it is relevant to inquire at this point
ics and cosm ology would have made such a bold idea unacceptable. About
whether the gradual acceptance o f the idea o f fluid rather than solid heavens
the time Philoponus’s works became available in the sixteenth century, the
played a significant role in the abandonment o f belief in celestial incorrup­
Copemican theory, which made the earth just another planet, was at hand
tibility. T ych o’s claim that the comet o f 1577 was m oving among the planets
to dissolve the medieval distinction between the terrestrial and celestial
clearly implied the nonexistence o f solid planetary spheres.100 For those who
regions. As the Copemican theory was gradually disseminated and adopted,
accepted comets as supralunar, a gradual but inexorable shift toward the
it became inevitable that the earth and its sister planets would be conceived
concept o f fluid heavens began. But did the idea o f fluid heavens imply
in terms o f the same matter.
corruptible heavens? A t least one Jesuit scholastic, Antonio Rubio, in a
What was strongly implied in Copernicus’s profound conceptualization
work o f 1615, believed that fluid heavens would have to be corruptible
was rendered graphically real by Galileo’s telescopic observations some
(presumably because o f divisibility) and therefore rejected them .101 But, as
sixty-five years after the publication o f De revolutionibus and only some
we have seen (n. 87 o f this chapter) others - for example, Amicus, Poncius,
fifteen to twenty years after Galileo had himself vigorously upheld the
and Oviedo - thought that the solidity or fluidity o f the heavens was ir­
incorruptibility o f the heavens and assumed the existence o f a celestial matter
relevant to the issue o f incorruptibility. Indeed, O viedo believed that the
that was radically different from its terrestrial counterpart. Beginning in
heavens were both fluid and incorruptible. For some scholastics, then, flu­
1610, however, with the publication o f the Sidereus nuncius (The Starry
idity alone did not necessarily entail divisibility. The matter o f the heavens
Messenger), Galileo rendered the centuries-old tradition o f celestial incor­
might be such that it was capable o f receiving only a single form, or celestial
ruptibility untenable.
matter might be incorruptible by virtue o f its form, a form that adhered to
Although Galileo did not explicitly declare the corruptibility o f the heav­
its matter so firmly that another could not be received.102 A seventeenth-
ens in the Sidereus nuncius, it was an obvious inference from the comparisons
century scholastic could therefore accept both fluidity and incorruptibility.
he made between the earth and the M oon, as when he observed that “ the
Although the shift from solidity to fluidity was a significant change from
terrestrial roughnesses are far smaller than the lunar ones” 103 and that the
the medieval tradition, it was not crucial for the issue o f celestial incorrup­
tibility. earth “ is movable and surpasses the M oon in brightness” and “ is not the
dump heap o f the filth and dregs o f the universe.” 104 In his Second Letter on
Sunspots, Galileo made the corruptibility o f the heavens explicit,105 and he
also argued for celestial corruptibility at great length in his Dialogue Con­
IV. Some concluding observations about celestial matter and cerning the Two C hief World Systems,'06 where, as evidence for corruptibility,
incorruptibility he cited comets, new stars, sunspots, and the M oon’s rough surface.107 A ll
the planets and stars were declared alterable by Galileo, even though the
From the various interpretations described earlier, it may be tempting to M oon and other planets might differ from the earth because their similar,
view Aegidius Romanus’s identification o f celestial and terrestrial matter as
103. Galileo, Sidereus nuncius [Van Helden], 1989, 51.
104. Ibid., 57-
100. Because the parallax o f the com ets placed them below the fixed stars, one could continue
105. Galileo, Letters on Sunspots (1613) in Drake, 1957, 118. The Second Letter on Sunspots is
to believe, as did Riccioli, that the fixed stars were embedded in a solid sphere. (Donahue,
dated A u gu st 14, 1612, and was printed in 1613.
1972, 117. holds that the sphere o f the fixed stars was the last element o f the old cosmos
to go.) 106. Galileo, Dialogue , First D a y [Drake], 1962, 4 1-10 0 .
101. Donahue, 1972, 105. 107. Ibid. 51, 72, 100. Galileo did not, however, believe that the M o o n ’s matter was like
102. O vied o, D e caelo, contro. i, punc. 2, 1640, 1:462, col. 1, par. 2. that o f our earth, and from this he inferred that i f plants and animals existed on the
M oon , they w ould be unlike those on earth.
270 THE CELESTIAL REGION

if not identical, matter was affected differently by the Sun and other en­
13
vironmental conditions.
Despite possible differences in lunar and earthly matter or even between
the earth and any other planet, Galileo w holly rejected the traditional con­
ception o f the earth's inferiority with respect to celestial bodies, an inferiority
The mobile celestial orbs:
that was based on the alleged immutability o f the heavens as contrasted to
the alterability o f terrestrial bodies. Through Sagredo, his spokesman in concentrics, eccentrics,
the Dialogue, Galileo posed a question, the answer to which would have
been assumed as self-evident in the Middle Ages. W hy should immutability
be more noble than mutability? “ For my part,” Sagredo replies,
and epicycles
I consider the earth very noble and adm irable precisely because o f the diverse al­
terations, changes, generations, etc. that o ccu r in it incessantly. If, not being subject
to any changes, it w ere a vast desert ot sand or a m ountain o f jasper, or if at the As we saw earlier (Ch. 1, Sec. II.2 and n. 12), the term caelum, which
time ot the flood the w aters w hich covered it had frozen, and it had rem ained an signifies heaven or heavens, was used to designate the entire celestial region
enorm ous globe ot ice w here nothing was ever born or ever altered or changed, I and occasionally even the spheres o f the elements below. Indeed, Thomas
should deem it a useless lum p in the universe, devoid o f activity and, in a w ord, Aquinas conceived all seven planetary spheres plus the sphere o f the fixed
superfluous and essentially nonexistent. T h is is exactly the difference betw een a stars as part o f a single “ sidereal heaven.” The customary usage o f the term
livin g animal and a dead one; and I say the sam e ot the m oon, o f Jupiter, and o f caelum was, however, more restricted and was intended to designate the
all other w orld globes. sphere or spheres that carried a single planet around, as when Galileo de­
clared; “ w e maintain that there are ten movable heavens [caelos mobiles], and
These words were written nearly one hundred years after the publication beyond these, that there is an eleventh, immovable heaven.” ' Although in
ot Copernicus s De revolutionibus. Despite the repudiation o f celestial im­ this chapter occasions will arise where the term “ heaven” (caelum) will be
mutability by Tycho in the late sixteenth century and by Kepler in the applied to a single orb or to the spheres o f a single planet, I shall more
second decade o f the seventeenth century. Galileo thought it necessary to regularly use the terms “ sphere(s)” or “ orb(s).” 2 The term “ heavens,” or
devote most o f the discussion o f the first day o f his Dialogue to that same “ heaven,” will continue to be used to designate the celestial region as a
theme. There his interpretations o f his ow n telescopic observations reduced whole.
the issue o f celestial incorruptibility to a nullity. After Galileo, scholastics,
as we have seen, were much less likely to assign incorruptibility as a property
o f celestial matter.10
8
I. O n e h eaven (or sphere) or m any?
108. Ibid., 58-59.
Despite the multiplicity o f planets and stars, it was not immediately obvious
to Aristotelian natural philosophers that the heavens were filled with in­
dependent orbs to which those celestial bodies were in some way attached
and which functioned to carry them around in their orbits. In his Metaphysics
(12.8.1074a.30-37 [Ross], 1984), Aristotle declares that “ there is but one
heaven.” For if more than one existed, matter would necessarily exist in
the heavens, because “ all things that are many in number have matter.”
But matter cannot exist in the heavens. Moreover, “ that which is moved
always and continuously is one alone; therefore there is one heaven alone.”
But in the same Metaphysics (12.8.1073b. 17-10743.16), immediately pre-

1. Galileo [De eaelo, qu. 1 (G)], 1977, 63. The Latin is from his Juvenilia, in Galileo. Opere
[Favaro], 1891-1909, 1:41.
2. On the largely ignored distinction between the terms “ orb” and “ sphere.” see Chapter 6.
note 37.

271
272 THE CELESTIAL REGION MOBILE CELESTIAL ORBS 273
ceding his declaration that the heavens are one, Aristotle describes and tures.” 7 Indeed, we cannot even be certain that we know the number o f
expands upon the famous astronomical systems o f Eudoxus and Callippus. planets, since other planets might exist that are invisible to us but which
N ot only did he increase the 33 spheres o f Callippus to 55, but he also might play a role in the generation o f sublunar things.8 In arriving at a
conceived o f the spheres as physical bodies, rather than as convenient math­ judgm ent about the oneness or multiplicity o f the heavens, scholastic natural
ematical constructions in the manner o f Eudoxus and Callippus. In De caelo philosophers frequently had to consider whether the celestial orbs were
Aristotle declares emphatically that each planet is carried around in its own continuous or contiguous, and the latter question occasionally involved opin­
sphere.3 Was there a conflict here? W ould the existence o f spheres divide ions about whether the celestial region was composed o f a fluid substance
the celestial region into a multiplicity o f heavens? And would a series o f or a series o f hard celestial spheres. Indeed, in the seventeenth century,
distinct spheres imply divisible, rather than indivisible, heavens? Both pos­ when “ solid” sphere was synonymous with “ hard” sphere (see Ch. 14, Sec.
sibilities were contrary to basic Aristotelian principles. VII), Roderigo de Arriaga declared that for those who assumed a fluid
In the sixth century, John Philoponus, an important Greek commentator heaven, the problem o f the number o f spheres was easy to resolve. They
on the works o f Aristotle but also a Christian, wrote a significant com­ would “ assume only one heaven [conceived as a single sphere] through
mentary on the six days o f creation. In it, he assumed that in Genesis Moses which the planets and stars move as [do] fish in the sea, or birds in the air.
was describing a single heaven, or firmament, which included all the planets The difficulty lies in the case in which the heavens are solid.” 9
and stars. Although astronomers were free to save the appearances by imag­ A t the outset, we must draw attention to an apparent ambiguity in the
ining as many orbs and motions as they wished, they had never demon­ medieval interpretation o f “ one heaven(s),” which was understood as syn­
strated the existence o f such entities. Indeed they did not themselves agree onym ous with the whole celestial region. Some conceived o f such a single
on the number o f these extra orbs and m otions.4 In his commentary on De heaven, or heavens, as devoid o f orbs, while others thought o f it as sub­
caelo, Averroes conceived o f the totality o f celestial orbs as a single heaven. divided into orbs that were either continuous or contiguous. B y contrast,
Indeed, he thought o f it as a single animal, whose partial orbs were like its some, perhaps most, assumed a celestial region that embraced numerous
members: that is, the motions o f the partial orbs were like the motions o f separate heavens in the form o f independent orbs that were regarded as
the members o f an animal. It was because o f this oneness that the whole contiguous to, and therefore distinct from, one another.
heaven could be moved with a single daily m otion.5 Those w ho believed in a single, physical heaven without orbs‘° under­
During the Middle Ages, scholastics frequently inquired whether the stood by this the entire celestial region, which embraced all o f the planets
heaven was one, continuous, indivisible body or a divisible entity comprised and stars and for some even extended down to the region o f the air. In this
o f a series o f distinct spheres, including one sphere for all the fixed stars system, the planets, which are not attached to orbs, are moved either by
and at least one sphere for each planet. Robert Grosseteste considered the themselves or by an intelligence. Although John Damascene and John C hry­
question on the oneness or multiplicity o f the firmament, which for him sostom may have adopted such an interpretation, and Robert Grosseteste
embraced everything from the M oon to the fixed stars, difficult.6 He con­ at least entertained it,11 an orbless heaven found few supporters during the
fessed an inability to answer it, asserting that “ no one can declare anything
certain about the number o f heavens or their motions or movers or na­ 7. “ N ullus potest de numero celorum aut eorum motibus aut motoribus aut ipsorum naturis
aliquid certum profiteri.” Ibid., part. 3, ch. 8, 3, 109, lines 7-9. Confessions ofign oran ce
about the firmament were not unusual. Durandus de Sancto Porriano [Sentences, bk. 2,
3. “ B ut the upper bodies [i.e., the planets] are carried each one in its sphere.” Aristode, D e dist. 14], 1571, 155V, col. 2, reports that Bede identified the firmament w ith the sidereal
caelo 2.7.2893.30 [Guthrie], i960. In his thirteenth-century translation, M ichael Scot ren­ heaven (coelum sydereum) that divides the waters from the waters, w hile som e thought
dered this passage as “ stellae autem, quae sunt in orbe superiori, procedunt in suo orbe.’ the firm am ent was air, others that it was water and yet others that it was fire. “ What
For M ich ael’s transladon in the edition o f A verroes’ commentaries, see Averroes [D e caelo, these are and for w hat they were created,” asks Durandus, “ are k n ow n on ly to the one
bk. 2, text 42], 15 6 2 -1574 . 5:124V, col. 2. w h o created [them]” (Quales autem sunt, et ad quid conditae, ille solus no'vit qui condidit).
4. D uhem , L e Systeme, 19 13-19 59 , 2 :1 1 1 -1 1 2 , 498—499, describes Philoponus’s interpretations 8. “ Sed unde scietur quod non sint plures stelle erratice nobis invisibiles, generacioni tamen
as he found them in the latter’s Greek text o f D e opificio mundi libri V II edited in 1897 by in inferiori m undo necessarie et utiles?” Hexaemeron, part. 3, ch. 8, 3, 108, lines 24-25.
Walter Reichardt. A s his source, D uhem gives bo o k 3, chapter 3, pages 1 1 3 - 1 1 6 . Tow ard T h e existence o f such planets could on ly be k n ow n b y revelation (“ U n d e igitur sciri
the end o f the fifteenth century, in Italy, G iovanni G ioviano Pontano (1429-1503) rejected posset, nisi a divina revelacione, an non sint plurime huiusm odi stelle invisibiles nobis,
the existence o f physical orbs and assumed that the orbs o f the astronomers were simply quarum quelibet suum habeat celum movens ipsam ad profectum generacionis in mundo
convenient devices to aid the understanding. H e believed that the planets were self-m oving. inferiori?” ). Ibid., lines 27-30.
See Trinkaus, 1985, 455. Even before T y c h o Brahe’s discoveries, the Jesuit Robert Bel- 9. “ N o v a hie Celebris occurrit de caelorum numero difficultas. Q u i caelos fluidos dicunt
larmine rejected the existence o f orbs in favor o f a fluid celestial m edium (see C h . 14. Sec. facile se ab hac quaestione expediunt: ponunt enim unum tantum caelum per quod planetae
VIII). astraque discurrunt, sicut pisces in mari aut aves in aere. Difficultas est casu quo caeli sint
5. See Averroes, De caelo, bk. 2, com m ent. 42, 1562-1574, 5:125V, col. 1. solidi.” Arriaga [D e caelo, disp. 1, sec. 4], 1632, 504, col. 1.
6. Grosseteste asks: “ Q u e sit autem huius firmamenti natura, et quot sint celi contend in hoc 10. Because o f its special nature, the empyrean heaven was always treated as a separate entity.
uno celo quod dicitur firm am entum .” Hexaemeron, part. 3, ch. 6, 1, 1982, 106, lines 1-2- 11. A lth o u gh John Damascene assumed that the firmament embraced all seven planets (he
274 THE CELESTIAL REGION MOBILE CELESTIAL ORBS 275
Middle Ages but came into vogue following upon Tycho Brahe’s famous they had to be carried around by gigantic orbs in which they were embed­
rejection o f hard, celestial orbs in favor o f a fluid, or soft, celestial region.12 ded.17 O n the further assumption that every motion attributed to a planet
The concept o f a single heaven divided into spheres was, however, not required its own unique orb, it followed that every orb moved with a single,
uncommon. Thus Thomas Aquinas assigned the seven planets and the fixed unique motion, in a single direction, but that different orbs could move in
stars to a single heaven, the “ sidereal heaven,” distinguishing eight spheres, diverse directions. The motion o f a single orb in a single direction could
one for each planet and the fixed stars.” Whether Thomas assumed a con­ not account for the astronomical phenomena.
tinuous heaven is uncertain, but Aegidius Romanus leaves no doubt that The need to account for the celestial phenomena was the reason why
he himself viewed the seven planets and the fixed stars as part o f one celestial orbs were deemed essential. They had been part o f the Greco-
continuous heaven that nevertheless contained deferents and eccentrics that Arabic legacy o f treatises that were translated into Latin in the twelfth and
were contiguous and therefore discontinuous. Aegidius thus insisted on a thirteenth centuries. More particularly, discussions about celestial orbs were
heaven that was continuous in one sense but discontinuous in another. (How embedded in the extensive Anstotelian-Ptolem aic body o f astronomical and
he presented this important and rather influential concept is described in cosmological literature. N o serious opposition to their acceptance appeared.
Section III.9.) More common, however, was Christopher Clavius’s inter­ Despite the assumption o f a multiplicity o f orbs, however, some scholastic
pretation o f a separate firmament o f fixed stars, under which there are seven authors conceived o f the celestial region as one continuous body, because
separate planetary spheres.14 thev assumed that the surfaces between successive spheres were continu-
Until the seventeenth century, when the effects o f T ych o ’s authoritative ous.,H Others denied such continuity and argued for the contiguity o f the
repudiation o f the existence o f solid celestial spheres began seriouslv to surfaces, and therefore for the distinctiveness o f each sphere. Still others
affect scholastic thought, almost all scholastic authors assumed the existence assumed a single, continuous heaven with diverse planetary channels or
o f real, physical spheres in the heavens.'5 The most fundamental reason for cavities that functioned as deferent or eccentric orbs. These interpretations
the postulation o f celestial orbs derived from the long-observed diversity depend on Aristotle’s definitions o f the continuity and contiguity o f surfaces.
o f planetary motions. Because planets were thought incapable o f self­ However, before defining these terms and examining the manner in which
movement and therefore unable to move through the celestial ether “ like thev were applied to the celestial orbs, we shall first describe the kinds ot
a bird through air or a fish through w ater,” as the popular phrase w ent,"’ spheres about which medieval cosmologists were concerned.
speaks ot “ the seven zones ot the firmament” [septem zonis tirmamenti], where each
zone contains a planet), he appears not to have assumed a division into orbs but rather II. C o n cen tric versus eccen tric orbs
into “ zones. He states, “ they say that there are seven zones ot the heavens, one better
than the other (Septem vero zonas aiunt esse caeli, aliam alia excelsiorem). See lohn
Damascene. De fide orthodoxa, 1955, 80, 82. Whether Damascene included the tixed stars 1. Aristotle's system o f concentric spheres
in the firmament is unclear, but Thomas Aquinas seems to have thought so (see Ch. 5,
Sec. VI.2). Between approximately 1160 and 1250, two rival cosmological systems
12. For example, Francisco de Oviedo ([De each, contro. 1, punc. 4], 1640. 471. col. 2), who entered western Europe and vied for acceptance. The first was derived from
rejected the existence o f hard orbs, held that a single, overall heaven could be conceived
as consisting ot several subheavens, “ because part ot a heaven through which the Moon the works o f Aristotle, where it was assumed that the stars and planets were
moves can be called the 'heaven ot the M oon,’ and another part through which the Sun carried around on, or in, concentric or homocentric spheres. As concentric
is moved can be called the ‘heaven ot the Sun,’ and the same for the other parts [of the spheres, they shared a common center, which was both the geometric center
heaven] through which the other planets are moved."
13. See Thomas Aquinas, Sinnma theohfiae. pt. 1, qu. 68, art. 4, 10:89. In the seventeenth o f the world and the center o f the earth. In this system, which he derived
century. Amicus included ail ot these in the tirmament. After arguing that “ the soliditv from his predecessors Eudoxus and Callippus, and which he describes all
that we proved applies to the tirmament. probably also applies to all the heavens o f the
planets, he declares that in Scripture "the term firmament not illy signifies the heaven orbs as fish are moved in water” (Aegidius Romanus, Opus Hexaemeroti. 1555, 49r, col.
o f the fixed stars but also [those] o f the wandering st.,rs.” Amicus [De eaelo, tract. 3, qu. 2). Duhem translates the section in Le Systeme, I9H-I959. 4014- One or the other of
5, art. 3], 1626, 279, col. 2. Amicus accepted the existence o f celestial orbs. the two descriptions appears in Conimbricenses [De eoelo, bk. 2, ch. 5, qu. 1, art. 1],
14. Clavius [Sphere, ch. 1], Opera. 1611, 3:23. M98, 246; Clavius [Sphere, ch. 4], 1593, 515; Aversa [De eaelo, qu. 32, sec. 6], 1627, 66,
15. As we saw in note 4. Pontano denied their existence. But Pontano does not qualify as a col. t; Amicus [De eaelo, tract. 5. qu. 4, dubit. 3, art. 1], 1626, 266, col. 1; Arriaga. De
scholastic author (see Trinkaus. 1985, 449). eaelo, disp. 1, sec. 3, subsec. 1. 1632, 499, col. 2; and Oviedo, De eaelo, contro. 1, punc.
16. Buridan declares that a planet can be imagined to be selt-moving “ bv dividing the orb 4. 164.0, 1:471. col. 2. For Robert Bellarmine’s use o f the metaphor, see Chapter 14,
itself just as a bird hies through air or a tish swims in water, or even as a man walks in Section VIII and note 76.
air.” See Buridan [De eaelo. bk. 2, qu. 18], 1942, 210-211. Many others used either the 17. Flow those orbs were thought to move is considered in Chapter 18.
bird or fish analogy. For example, see Aegidius Romanus’s discussion ot eccentrics and 18. For example, Saint Bonaventure [Sentences, bk. 2, dist. 14, pt. 2, art. 1. qu. 1], Opera.
epicycles, where he declares that it is unreasonable to sav that “ the planets are moved in 2:332. col. 1.
276 THE CELESTIAL REGION MOBILE CELESTIAL ORBS 277
too briefly in book 12, chapter 8, o f the Metaphysics,19 Aristotle assumes Aristotle fails to mention or further explicate the unrolling spheres anywhere
the existence o f 55 spheres. O n the basis o f his study o f Eudoxus and else in his w orks, a close examination o f what the real physical relations o f
Callippus, Aristotle assigned 33 concentric spheres to account for the mo­ these spheres might be yields little but frustration and confusion.22 It is with
tions o f the seven planets. O f these 33 spheres, 4 each were assigned to good reason that Dicks (1970, 203) has suggested that Aristotle may have
Saturn and Jupiter and 5 each to Mars, Venus, Mercury, the Sun, and the considered it only “ an interesting speculation, but one that would not stand
Moon. Then Aristotle declares (ibid., 1073^38-10743.5): up to close scrutiny.”
If Aristotle was spare in the description o f his system o f concentric
It is necessary, if all the spheres put together are going to account for the observed spheres, his Greek commentators - Sosigenes, Simplicius, and Philoponus,
phenomena, that for each of the planetary bodies there should be other counteracting for example - were more lavish and made some effort to explain Aristotle’s
[literally “ unrolling” ] spheres, one fewer in number [than those postulated by Cal­ meaning and intent. From the thirteenth to the seventeenth century, Ar­
lippus for each set] and restoring to the same function each time as regards position istotle’s system and number o f concentric orbs were rarely discussed in
the first sphere of the planetary body situated below; for only thus is it possible for questions on the Metaphysics, although they did receive consideration, if
the whole system to produce the revolution of the planets.20 only summarily, in section-by-section commentaries on the Metaphysics,2}
N ot only were commentaries on the Metaphysics rarer than questions on
Thus did Aristode assign 22 additional “ unrolling” spheres, for a total the Metaphysics, but the difficulty and obscurity o f Aristotle’s purely con­
o f 55. D. R. Dicks (1970, 200-201) has described the relationship between centric system tended to discourage full-blown discussions. Indeed, it was
the regular and unrolling spheres o f Saturn as follows: supplemented, if not largely supplanted, by a second system that also
reached the Latin West in the great wave o f Greco-Arabic translations o f
Thus for the four spheres of Saturn A, B, C, D, a counteracting sphere D' is
the twelfth and thirteenth centuries.
postulated, placed inside D (the sphere nearest the earth and carrying the planet on
its equator) and rotating round the same poles and with the same speed as D but
in the opposite direction; so that the motions of D and D' effectually cancel each 2. Ptolem y’s system o f eccentric spheres
other out, and any point on D will appear to move only according to the motion
of C. Inside D' a second counteracting sphere C ' is placed, which performs the In the second system derived ultimately from Claudius Ptolem y’s Hypotheses
same function for C as D' does for D; and inside C ' is a third counteracting sphere of the Planets, the planets were assumed to be carried around by a system
B' which similarly cancels out the motion of B. The net result is that the only o f material eccentric and epicyclic spheres, whose centers were geometric
motion left is that of the outermost sphere of the set, representing the diurnal points that did not coincide with the earth’s center.24 Despite its momentous
rotation, so that the spheres of Jupiter (the next planet down) can now carry out role in medieval cosmology, the Hypotheses o f the Planets was not translated
their own revolutions as if those of Saturn did not exist. In the same manner, Jupiter’s into Latin during the Middle Ages. In some as yet unknown manner, how ­
counteracting spheres clear the way for those of Mars, and so on (the number of ever, its fundamental ideas reached western Europe, probably in works
counteracting spheres in each case being one less than the original number of spheres translated from Arabic. The precise treatise, or treatises, in which these
in each set) down to the Moon which, being the last of the planetary bodies (i.e. ideas were embedded have yet to be identified. Although works attributed
nearest the earth), needs, according to Aristotle, no counteracting spheres. to Alhazen (ibn al-Haytham), Alfraganus (al-Fargani), and perhaps Thabit
ibn Qurra included descriptions o f material eccentric and epicyclic spheres,
This is indeed the extent o f Aristotle’s “ detailed” account o f the number
and interrelationships o f the concentric spheres that carry the planets around 22. D icks (1970, 203) declares that “ i f the heavens really operated in this, manner, w ith the
counteracting spheres effecting their respective cancellations o f planetary motions, how
the sky. N ot only is it exceedingly brief, to the point o f obscurity, but did astronomers ever m anage to m ake the obervations that lay behind the original Eudoxan
Aristotle even left scholars in confusion over the actual number o f spheres, schem e w ith its planetary loops and retrogradations?”
suggesting that if one subtracts certain motions from the Sun and Moon, 23. Jean Buridan’s Questions on the Metaphysics lacks such a discussion, while Thom as Aquinas’s
Commentary on the Metaphysics includes one (Thomas Aquinas [Metaphysics, bk. 12, lesson
the total number o f spheres would be 49 (10743.12-14). Some have argued 10], 1961, 2:904-907).
that 55 is required, while others reject 55 and argue for 49 or 61.21 Because 24. In this section, I rely on Grant, 1987b, 189 -19 5. D uhem , L e Systeme, 19 13 -19 5 9 , vols.
3 -4 , describes the controversy between the defenders o f concentric orbs and those w ho
19. Indeed, this is the on ly place am ong the currently extant w orks o f A ristotle in which he sided w ith Ptolem y and assumed eccentric and epicyclic orbs. D uhem chose to treat this
provides any details about his system o f concentric spheres. im portant topic b y taking up a series o f individual authors in chronological order. T h e
20. T h e translation is by D . R. Dicks, 1970, 200. concentric-eccentric controversy was but one o f a number o f themes discussed for each
21. Dicks accepts 55 (1970, 202-203), w hile Hanson (1973, 69) argues for 49 or 61, with the author. Sections o f varying lengths are thus isolated w ithin descriptions o f individual
latter preferable. authors and are nowhere adequately summarized or synthesized.
278 THE CELEST IAL REGION M O BIL E CELESTIAL ORBS 279

the basic scholastic version ot eccentrics and epicycles described in this the most important medieval description ot eccentrics and epicycles, as these
chapter finds no counterpart in the Latin translations o f these Arabic trea­ were understood by natural philosophers, forms the basis o f this discussion.
tises.2'' Because Bacon and subsequent authors regularly assumed at least three
References to epicycles and eccentrics appear in widely used thirteenth- eccentric orbs tor each planet,29 I shall frequently refer to the “ modern"
century works like Sacrobosco’s On the Sphere and in the anonymous Theo- theory as the "three-orb system ,” but I shall also refer to it as the "A ris-
rica planetarum, although neither author implies or suggests that they might totelian-Ptolemaic system,” since Aristotle’s concentric spheres were as­
be real, material, solid orbs.226 It Roger Bacon (ca. 1219—ca. 1292) was not
5 signed a significant role within the system o f eccentrics.
the first to mention material eccentrics and epicycles in the Latin West, he
may well have been the first scholastic natural philosopher to have presented
j. The system o f eccentrics: Roger Bacon
a serious evaluation o f their cosmological utility.27 After some hesitation
and ambivalence. Bacon rejected physical eccentrics and epicycles and opted After demonstrating the impossibility o f any system comprised o f eccentric
for Aristotle's system o f concentric spheres. Ironically, it was his description orbs in which the eccentricity is due to an eccentric convex surface, an
o f the system o f eccentrics and epicycles that was most widely adopted by eccentric concave surface, or the eccentricity ot both surfaces. Bacon intro­
medieval natural philosophers and whicn still found defenders well into the duces another interpretation - “ a certain conception of the moderns,” as he
seventeenth century. Although many scholastics ignored the topic o f ec­ put it3° - in which the external surfaces ot each planetary orb are concentric
centrics and epicycles, those who gave it more than a cursory glance in­ but which contain at least three eccentric orbs. To illustrate the system,
cluded, in the thirteenth century, Albertus Magnus and Duns Scotus; in the Bacon describes the motions o f the Sun and Moon. Because the Sun has
fourteenth century Aegidius Romanus, John o f Jandun, Jean Buridan, and only an eccentric orb, it will be useful to follow his account o f the Moon,
Albert o f Saxonv; in the fifteenth century Pierre d’A illy,28 Cecco d’Ascoli, which has both an eccentric and an epicyclic orb.
Johannes de Magistris, Johannes Versor, and Thomas Bricot; in the sixteenth In the diagram (Fig. 8),31 let T be the center o f the earth and world and
1. cntury John Major, the Coimbra Jesuits, and Christopher Clavius; and in also the center o f the lunar orb. The entire sphere o f the M oon lies between
the seventeenth century Bartholomew Amicus. Bacon’s account, perhaps the convex circumference A D B C and the concave circumference O Q K P,
which are both concentric to T. Between these two circumferences, three
25. Although it has been said (Pedersen [Lindberg], 1978, 321 and then 3 19) that the machinery orbs are distinguished (namely a', b', and c') by assigning another center,
o f the material spheres suggested by Ptolemy's Hypotheses of the Planets was presented to V, toward the M oon’s aux, or apogee. Around V as center are two cir­
Western astronomers in two brief cosmological treatises by Thabit ibn Qurra, namely
Thabit's De hiis quae indigent antequam legatur Almagesti and De quantitatihus stellarum et cumferences, A G F E and H N KM , which signify the surfaces that enclose
planetarum et proportio terrae, an examination o f these treatises reveals nothing relevant for the lunar deferent and form the eccentric orb b’ . Surrounding the eccentric
the problem ot material or physical eccentric and epicyclic spheres (for the texts, see orb or deferent is the outermost orb, a', lying between surfaces A D B C and
Thabit ibn Qurra jCarmodv], i960, 131-139, 145-148). What Thabit may have passed
on to the West on the subject o f material eccentric spheres will be seen later in this chapter. A G F E ; and surrounded by the eccentric orb is the innermost orb, c', lying
26. For Sacrobosco s discussion, see Sacrobosco, Sphere, 1949, 113-114 (Latin), 140-141 (En­ between the concave surface H N K M and the convex surface O Q K P . Be­
glish). For the Theorica, see Olat Pedersen’s translation in Grant, 1974, 452-465 (eccentrics tween the surfaces o f the middle, or eccentric, orb is a concavity that contains
are defined on page 452). In speaking ot “ real, material, solid orbs,” I deliberately refrain
from signifying whether those orbs were conceived as hard or soft, a problem that will a spherical epicycle. The latter may be conceived in two ways: either as a
be considered later. solid globe, which Bacon calls a "convex sphere” (spericum connexion), which
27. Sometime in the 1260s, Bacon presented almost identical accounts in his Opus tcrtlum and resembles a ball (pila) because it lacks a concave surface; or as a ring with
in the second book o f his Communia tiaturalium, or the De celestihus, as I shall refer to it.
In the latter. Bacon added a significant chapter on whether eccentrics and epicycles were two surfaces, one convex (K L F I), the other concave (R Y S ©), wrhere the
consistent with Aristotelian cosmology and followed this with a lengthy description o f
the Ptolemaic system. See Bacon, Opus tertium. 1909, where Bacon's discussion on ec­ 29. See Duhem, Le Systeme, 1913—1959, 4:112.
centrics and epicycles extends over pages 99—137. For the De celestihus, see Bacon, Opera, 30. “ De quadam ymaginatione modernorum." See Bacon. Opus tertium, 1909, 125 (and 125-
fasc. 4, 1913, 419-456. A. C. Crombie and J. D. North believe the Opus tertium was 134 tor the exposition and critique ot' the modern theory) and Bacon, De celestihus. fasc.
written sometime between 1266 and 1268 (see their article on Bacon in Dictionary of 4, 1913, 438 (and 437-443 tor the modern theory). Although the “ modern” theory seems
Scientific Biography, 1970-1980, 1:378) and suggest that Bacon may have written his Com­ ultimately derived from Ptolemy’s Hypotheses of the Planets, Bacon's immediately pre­
munia naturalium in this same period. By contrast, Lindberg [Bacon], 1983, xxxii, who ceding discussion (Opus tertium, 119—125; De celestihus, 433-437), concerning the impos­
accepts 1267 as the date o f composition for the Opus tertium, believes that no firm date sibility o f a total planetary orb being composed o f eccentric orbs where one or both of
can be attached to the Communia naturalium and concludes: “ all that can be said is that it the external surfaces is eccentric, may have derived from an earlier attempt to materialize
represents the early stages ot the broadening o f Bacon's outlook, the usual guess placing eccentrics on the basis o f Ptolemy’s description in the Almagest.
it in the earlv 1260s.” 31. The figure appears in Bacon, Opus tertium, 1909, 129; the relevant text is on pages 128-
28. If d'Ailly’s treatment was perhaps less extensive than Bacon's, it was nevertheless lengthier 13 1. The figure was used in Grant, 1987b, 191. To identity the three distinct orbs, ! have
than most others. added the letters a', h‘ , c'.
280 THE CELESTIAL REGION MOBILE CELESTIAL ORBS 281

thirteenth, and even into the fourteenth, century, there were scholastic
natural philosophers w ho refused to embrace, without qualification, a world
constructed almost w holly o f eccentric orbs. In a number o f places in his
writings, Thomas Aquinas was either noncommittal (De trinitate) or rejected
eccentric orbs (Commentary on the Metaphysics). In his last treatise (De caelo),
he argued that the existence o f eccentrics and epicycles was undemonstrated,
because even though they are useful for saving the astronomical appearances,
they might not be physically real.3s Although John ofjandun was convinced
that the three-orb system could save the astronomical appearances - indeed,
he proclaimed that he knew o f no argument that could repudiate it - he
rested content in the end to proclaim it as merely “ possible.” 36
Such reservations and hesitations were, however, more and more the
exception by the end o f the thirteenth and the early fourteenth century.
Scholastics increasingly came to assume the existence o f real, material ec­
centric orbs, as did, for example, Albertus M agnus,37 Duns Scotus, Aegidius
Romanus (ca. 1245-1316), and Durandus de Sancto Porciano (d. 1334).

4. The system o f eccentrics: Pierre d ’Ailly

Figure 8. Representation of the Moon’s concentric, eccentric, and epi- By the end o f the fourteenth century, some 130 years after Bacon’s account,
cyclic orbs as described in Roger Bacon’s Opus tertium. (Diagram from the three-orb system had received its definitive scholastic form. N o one
Bacon, Opus tertium, 1909, 129.) expressed it better than Pierre d’A illy, who presented as detailed an account
as could be expected from a natural philosopher who was not a technical
central core belongs w holly to the eccentric orb and forms no part o f the astronomer. Keeping in mind Figure 8, it will be useful to sketch d’A illy’s
epicyclic sphere itself. The M oon (luna in Figure 8), or planet, is a solid description o f the three-orb system as he presented it in his 14 Questions on
spherical figure which has only a convex surface and is located within a the Sphere o f Sacrobosco,38 N ot only does he em ploy a technical terminology
concavity o f the epicyclic orb. The eccentric sphere is assumed to move that is largely absent in Bacon, but the objections he raises and the solutions
around its center, V, carrying the epicycle with it; the epicyclic sphere, in he proposes were representative o f the way material eccentrics and epicycles
turn, has its own simultaneous motion and carries the planet with it. I shall were interpreted by most natural philosophers from the late Middle Ages
first describe medieval concern for eccentrics and then consider epicycles. to the end o f the sixteenth century.
When extended to all the planets, it was this system that was widely According to d’Ailly, the heavens are made up o f a combination o f con­
adopted during the late Middle Ages. Even those w ho did not accept the centric and eccentric orbs. The totality o f every sphere or orb (orbis totalis)
three-orb system believed it saved the astronomical appearances better than is concentric and includes within it all other orbs necessary to produce the
did the systems o f concentric spheres proposed in Aristotle’s Metaphysics position o f the planet. Within the concentric surfaces o f each planetary orb
(I2.8.i0 73b.n-I074a. 14) and in al-Bitrujl’s more technical De modbus ce- are the eccentric orbs. Each eccentric orb or sphere, usually described as a
lorum.32 N ot even Averroes’ strong support for Aristotle’s concentric as­ “ partial orb” (orbis partialis),39 contains the center o f the world as well as
tronomy and cosm ology33 could entice medieval natural philosophers from
35. For these passages, see Litt, 1963, 348, 350-352. T h o m as’s position is akin to that o f
the conclusion that Ptolemaic eccentric orbs were superior for saving the
M oses M aim onides (see M aim onides, Guide [pt. 2, ch. 24], 1963, 2:322-327).
astronomical phenomena. Indeed, only a few unambiguous defenders of 36 . John o fja n d u n [Metaphysics, bk. 12, qu. 20: “ W hether a plurality o f eccentric orbs and
Aristotle’s purely concentric cosm ology can be identified.34 But during the epicycles is really in celestial bodies” ], 1553, I4 ir, col. i - i4 2 r , col. 1.
37. A lth o u gh Albertus M agn us accepted the existence o f eccentrics, his arrangement o f them
32. T h e Latin text appears in Bitrujl [Carm ody], 1952. differed from B acon ’s popularly accepted description.
33. Averroes’ defense o f Aristotle was made in his middle com m entaries on D e caelo and the 38. Pierre d’ A illy devoted the thirteenth o f his fourteen questions on the Sphere o f Sacrobosco
Metaphysics. T h e relevant passages have been collected and analyzed b y C arm o d y, I 95 2- to the question “ Whether it is necessary to assume eccentric and epicyclic circles to save the
556-586. appearances in planetary m otions.’’ See d’A illy, i4Q uestions, qu. 13, 1531, 163V-164V.
34. For exam ple, W illiam o f Auvergne, Alexander o f Hales, and Saint Bonaventure. See 39. A lth o u g h the expression orbis partialis was rather com m on, d ’A illy did not use it. Clavius,
Duhem , L e Systeme, 19 13 -19 5 9 , 3:404. Sphere, ch. 4, 1593, 502, how ever, did use and define it, saying that a w hole sphere was
282 THE CELESTIAL REGION MOBILE CELESTIAL ORBS 283

its own proper center that is eccentric with respect to the center o f the With these definitions and concepts, d’Ailly describes next the three-orb
world. scheme in a manner that differed little from Bacon’s earlier description o f
Like Bacon, d’Ailly distinguished th,ree types o f eccentric orbs but divided the theory o f the “ moderns.” 44 According to d’A illy, astronomers imagine
them into two classes: one, called eccentricus simpliciter, has the same center three eccentric orbs as constituting the whole sphere o f a planet. T w o o f
for both its concave and convex surtaces; the other, designated eccentricus these orbs are eccentrics secundum quid, that is eccentric with respect to one
secundum quid, has the center o f the world as the center o f one surface and surface only. One o f them is the outermost orb and the other the innermost
a point outside the center ot the world as the center o f the other surface. orb. As eccentrics secundum quid, the outermost orb is eccentric only with
The former surface is concentric, the latter eccentric. Thus the eccentric respect to its concave surface, while the innermost orb is eccentric only
surface o f an eccentric orb secundum quid may be either convex or concave, with respect to its convex surface. Between these tw o orbs lies the third,
yielding two different types ot eccentric orbs for a total o f three.40 Because which is eccentricus simpliciter, because both o f its surfaces are eccentric.
it has two eccentric surtaces, an eccentricus simpliciter will always be o f uni­ Indeed, the middle eccentric is constituted ot the concave surface o f the
form thickness and is called the “ deferent” orb because it carries the planet. outermost eccentric sphere and the convex surface o f the innermost sphere.
The deferent orb is divided into four equidistant points: the aux, which is The middle orb is called the “ deferent orb” and carries the planet itself.
most distant trom the center o f the world; the opposite o f the aux, which With the exception o f the outermost and innermost orbs o f the world, each
is the point on the deferent nearest to the center o f the world; and the twro orb was conceived as a ring-like figure that contained other ring-like orbs
opposite points, located between the aux and opposite o f the aux, are called and was contained by other ring-like orbs.43 Raphael Aversa provides further
the mean distances (lonqitudities mediae).4' B y contrast, the eccentricus secundum details when he explains that
quid, with one surface concentric and the other eccentric, is thicker in one
part and thinner in another.4" When eccentric orbs are moved, the thin part the planet itself, w hich is carried in such an orb [i.e., in the eccentric deferent], n o w
o f one moves with the thick part o f another and conversely.43 arrives o ver the thicker part, n o w o ver the thinner part o f the extrem e lo w e r orb.
A n d so for h a lf o f its path it recedes a bit from the earth, and in the other h a lf it
composed ot partial orbs (orbes partiales). Philip Melanchthon (1550, 52V) also spoke of draw s nearer [to the earth]. A n d it also takes m ore tim e in the thicker h a lf than in
the geometers who fashion three partial orbs for the Sun (“ Solis tres partiales orbes” ). the thinner half, because that part o f its circuit is greater. T hus it seems to turn
40. Without employing the terms eccentricus simpliciter and eccentricus secundum quid. Albert of
Saxony ([De celo, bk. 2, qu. 7: “ Whether for saving the appearances of the planetary m ore slo w lv then. [Finally], w hen it is in the sum m it o f the thicker part, it is said
motions, it is necessary to assume eccentric orbs and epicycles” ], 1518, io6v, col. 1) to be in the aux: and w hen it is in the thinner part it is said to be in the opposite
divided eccentric orbs into the same three types as did Bacon and d’Aillv. Because d’Aiily's o f the aux. T h e aux is called apogee; [the] opposite [o f the aux is called] p erigee.40
arguments are similar to. and even follow the order of, Albert's, it is not unreasonable
to suppose that d’Ailly may have used Albert’s question as one o f his chief sources.
Among authors who wrote after d’A illy, Johannes de Magistris [De celo. bk. 2. qu. 3],
5. E p ic y c le s
1490, 21, col. 1. expressed the same threefold distinction when he said, "there is a certain
eccentric with respect to each surface [i.e., its convex and concave surtaces are eccentric];
another [kind] is eccentric with respect to only one surface [i.e., either to the convex or
Although the deferent orb carries the planet, it does so by means o f an
concave surface]” (Quidam est eccentricus secundum utramque superficiem. alius est epicycle, thought o f as either a solid globe or a ring (and depicted in the
eccentricus secundum unam tantum). Thomas Bricot repeats the same xiea (Bricot [De
celo. bk. 2], i486. 2iv, col. 2), as does John Major ([Sentences, bk. 2. dist. 14, qu. 4), 2, dist. 14, qu. 4, 1519b. 7sv, cols. 1-2, seems to say much the same thing, when he
1519. 75V. cols. 1—2). In the late sixteenth century. Clavius spoke o f the same two classes declares that '‘the differences o f thickness are so arranged that the thicker part of one
o f eccenuics and used the same terminology as d’Ailly, as did Aversa, Amicus. Mastrius ahvays corresponds to the thinner part o f another” (Sunt dittormes spissitudims sic ordmati
and Bellutus, and Cornaeus in the seventeenth century. See Clavius, Sphere, ch. 4, 1593, ut semper parti spissiori umus pars tenuior alterius respondeat). Aversa, De caelo. qu. 52.
499; Aversa, De caelo, qu. 32. sec. 5, 1627, 58, col. 2; Amicus, De caelo, tract. 5, qu. 4, sec. 5, 1627, 58, col. 2. and Mastrius and Bellutus, De coelo, disp. 2. qu. 1, art. 2, 1727,
dubit. 3, art. 3, 1026. 267, col. 1; Mastrius and Bellutus [De coelo. disp. 2, qu. 1, art. 2], 3:488, col. 2, par. 22, also held the same opinion.
1727, 3:488. col. 2, par. 22; and Cornaeus [De coelo, disp. 2. qu. 2. dub. 3], 1657, 494- 44. With this proviso: d’Ailly's range o f technical terminology was much greater.
495. As we shall see. ot chis group only Cornaeus rejected eccentrics and epicycles une­ 45. In the seventeenth century, Aversa gave a thorough summary o f the three-orb system
quivocally. along the same lines and with much the same terminology in De caelo, qu. 32, sec. 5,
41. The same four points are mentioned by Major [De celo. bk. 2, qu. 2], 1526, 15. 1627, 58, col. 2-60, col. 2. For d’A illv’s description o f an orb as a ring-like figure with
42. Although the diversity ot thickness in eccentrics secundum quid was obvious, it was ex­ a convex and concave surface, see my discussion in Chapter 6, note 37.
plicitly mentioned by some: for example. Clavius, Sphere, ch. 4, 1593. 499; Aversa. De 46. “ Unde sit ut ipsum astra, quod in tali orbe defertur, modo incedat super partem cras-
caelo, qu. 32, sec. 5. 162-. 58, col. 2-59, col. 1; Amicus, De caelo, tract. 5, qu. 4, dubit. siorem, modo super tenuiorem orbis extremi inferioris. Et sic pro medietate sui circuitus
3, art. 2, [626. 266, col. 2: and Cornaeus, De caelo, disp. 2, qu. 2, dub. 3, 1657, 495. paulatim altius a terra recedat: pro altera medietate proprius accedat. Atque etiam plus
43. See d’Aillv, 14 Questions, qu. 13, 1531. 163V. Duns Scotus agreed when he declared that tempons connciat in medietate crassiori quam tenuiori quia inaior est ilia pars sui circuitus.
“ the thicker part o f one lies against the thinner part o f another, and conversely” (Semper Et ita pro tunc videatur tardius gyrare. Atque dum est in summo partis crassions dicitur
enim spissior pars umus est contra partem minus spissam alterius, et e converse). Duns esse in auge; dum est in imo partis subtilions dicitur esse in opposito augis. In auge dicitur
Scotus [Sememes, bk. 2, dist. 14, qu. 2], Opera, 1639, 6, pt. 2:732. Major, Sentences, bk. apogaeum; in opposito perigaeum." Aversa. ibid., 59, col. 1.
284 THE CELESTIAL REGION MOBILE CELESTIAL ORBS 285

diagram as the circle KLFF). D ’A illy describes the epicycle as “ a small circle - namely the surface o f the outermost eccentric orb and the concave surface
on the surface o f the deferent orb that does not contain within itself the o f the innermost eccentric orb - had to be concentric. This was, o f course,
center o f the world; and the body o f the planet is imagined to be in it. And a primary feature o f the Aristotelian-Ptolemaic compromise that saved the
this epicycle is assumed to be contiguous, and not continuous, with the geocentric system. Because a concentric orb has the geometric center o f the
eccentric deferent because it is moved with a motion other than the motion world as its center, d’A illy explains that “ the first movable sphere [the
o f the eccentric deferent.” 47 Like the eccentric deferent, the epicycle has primum mobile, sometimes equated with the sphere o f the fixed stars] is a
four equidistant points: the aux o f the epicycle; the opposite o f the aux o f concentric orb, and generally every total orb is concentric, where ‘total orb’
the epicycle; and two points equidistant from the aux and the opposite of [orbis totalis] is taken as the aggregate o f all the orbs required to save the
the aux called “ stations” (stationes), one o f which marks the point at which total motion o f a planet.” 50 In this manner, Aristotle’s cosm ology o f con­
the planet begins its retrograde motion (point I o f the epicycle in Figure 8), centric spheres was saved, even though, in violation o f his physical prin­
the other o f which marks the point where it begins its direct motion (point ciples, eccentric orbs with centers other than the center o f the world formed
L in Figure 8).48 the basis o f the compromise system.
Like eccentric orbs, epicyclic orbs represent the various observed motions Here, then, was the compromise that produced the three-orb system in
and dispositions o f the planets, which, according to d’A illy, are o f three which three partial eccentric orbs are encompassed within tw o concentric
types: direct motioQ, retrogradation, and station. When a planet is in the surfaces that together form a single concentric orb (see Figure 8). The latter
aux o f its epicycle, its motion is said to be direct and quickest, because the was then perceived as representing the total motion o f the planet. The three-
direction o f its motion on the epicycle is the same as that o f the eccentric orb system was forced upon Aristotelian natural philosophers because A r­
deferent. But when the planet is in the opposite o f the aux o f the epicycle, istotle’s straightforward concentric system could not account for the astro­
its motion is retrograde and slower, because it now moves in a direction nomical phenomena. Thomas Bricot observed that there are numerous
opposite to that o f the eccentric deferent. Should the planet arrive at one irregularities in the celestial motions, but not all require the assumption o f
o f the points o f station, it would m ove neither with the deferent nor contrary eccentrics or epicycles.51 But one irregularity that definitely requires an
to it, so that its speed will seem neither to increase nor decrease and it will eccentric is the variation in a planet’s distance from the earth.52 It was the
appear stationary.49 most fundamental reason w hy almost all scholastic natural philosophers felt
compelled to accept the compromise and abandon Aristotle’s purely con­
centric system. They consoled themselves with the thought that the total
6. The great compromise: the three-orb system
orb (orbis totalis), which contained the three eccentrics, was concentric and
Although d’A illy’s three-orb system incorporates the three types o f eccentric that in some sense one could still speak o f a concentric world system. The
orbs that he, Bacon, and many others distinguished, the enclosing surfaces difficulty in all this was the fact that although the earth remained the center
o f the concentric orbs, it could not function as the center o f the eccentric
47. D ’A illy, 14 Questions, qu. 13, 1531, i6 3 v -i6 4 r . For the position o f the epicycle, see Figure
8. Sacrobosco defines an epicycle as “ a small circle on w hose circumference is carried the orbs. Because most scholastics could see no plausible alternative, it was an
bo dy o f the planets, and the center o f the epicycle is always carried along the circumference anomaly they had to accept. That is why, although d’A illy mentions rival
o f the deferent.” Sacrobosco, Sphere, ch. 4, 1949, 141 (Latin text, 114). M ajor, Sentences,
interpretations which rejected epicycles and eccentrics, he regards the Pto-
bk. 2, dist. 14, qu. 4, 1519b, 75V, col. 2, also says that the epicyle is contiguous, rather
than continuous, w ith its deferent, “ because it is m oved w ith another m otion than the
m otion o f the deferent” (quia m ovetur alio m otu quam m otu deferentis).
48. D ’A illy, ibid., i64r. B rief definitions o f the points o f station appear in Sacrobosco’s Sphere 50. “ U n d e orbis concentricus diritur orbis sub utraque eius superficie continens centrum
(ch. 4, 1949, 141) and in the Theorica planetarum (for both, see Grant, 1974, 450 and 461, m undi et habens eius centrum cum centro mundi. Isto m odo prim um m obile est orbis
respectively). Major, Sentences, bk. 2, dist. 14, qu. 4, 1519b, 75V, col. 2, identifies the
;i| concentricus et generaliter quilibet orbis totalis est concentricus et ibi capitur orbis totalis
same four points. H ow ever, the four points on the epicycle are in fact not equidistant, pro aggregato ex om nibus orbibus requisitis ad salvandum m otum totalem unius pla-
as d ’A illy and M ajor assert. A lth o u gh the first and second stations m ust alw ays be equi­ netae.” D ’A illy , 14 Questions, qu. 13, 1531, 163V.
distant from the true apogee (aux) o f the epicycle, the points o f station alw ays fall nearer 51. H e describes five such irregularities, or dififormities (difformitates planetarum), and declares
to the perigee o f the epicycle (opposite o f the aux) than to the apogee. It follow s therefore that “ such irregularities can be saved b y the assumption o f several m otions in the same
that the four points on the epicycle m entioned b y d’A illy cannot be equidistant. See celestial b o d y w ithout the assumption o f eccentrics and epicycles” (Et tales difformitates
Cam panus o f N ovara, Theorica planetarum, 19 71, 225-227, 231, 313. I am grateful to my possunt salvari per positionem plurium m otuum in eodem corpore celesti sine positione
late colleague, V ictor E. Thoren, for bringing this to m y attention, eccentricorum vel epiciclorum). Bricot, D e caelo, bk. 2, i486, 29r, col. 2. A m o n g the
49. In the thirteenth century, B artholom ew the Englishm an provided an earlier descriptive irregularities, Bricot includes a planet’s proper m otion, w hich is contrary to its daily
version o f eccentrics and epicycles (bk. 8; 1601, 398-399). A lth o u gh he speaks o f equant, m otion; m ovem ent along the zodiac from tropic to tropic; and variations in latitude with
deferent, and epicyclic circles and also o f direct and retrograde m otion, as w ell as or respect to the ecliptic.
station, B artholom ew ’s account lacks the more sophisticated term inological distinctions £ i 52. “ Sed est alia difformitas in approprinquatione planete ad nos et in elongatione eius a nobis.
found in d’ A illy ’s treatise. E t talis non potest salvari sine eccentricis, sed bene posset salvari sine epiciclis. ” Ibid.
286 THE CELESTIAL REGION M O BILE CELESTIAL ORBS 287

lemaic theory o f eccentrics and epicycles as “ more com m on” (est magis situations and to reconcile eccentrics and epicycles with Aristotelian cos­
communis) and unambiguously adopts it. m ology and physics.
Until the seventeenth century, most.scholastic authors assumed the truth In what follows, I shall describe the manner in which medieval natural
o f the three-orb compromise. In that century, however, Tycho Brahe’s philosophers coped with materia] eccentrics and epicycles within the frame­
analysis o f the comet o f 1577, which indicated the nonexistence o f solid, w ork o f Aristotelian cosmology and physics. An early account — probably
hard spheres o f any kind, began to make its influence felt. The gradual one o f the first - o f the problems inherent in a system o f eccentrics and
acceptance o f a fluid celestial region led even some scholastic authors to epicycles appears in Vincent o f Beauvais’ Speculum naturale, where, some­
abandon the idea o f eccentric and epicyclic spheres. But the traditional time around 1244,34 in a chapter titled “ Whether there is any space between
opinion continued to hold the allegiance o f some important scholastics in the spheres o f the planets,” ” Vincent asked whether there is any body or
the late sixteenth and the seventeenth century, including Clavius, Amicus, void space between spheres or whether the spheres mutually touch.5* Citing
Mastrius and Bellutus, and perhaps A versa.’3 O f this, however, more will an author whom he calls “ Avenalpetras,” who is probably the Arabian
be said when we consider scholastic opinions on the hardness or fluidity o f astronomer al—BitrujI, ” Vincent relates that Avenalpetras assumed that the
the celestial region. N ow , however, we must examine the physical and celestial orbs are in contact and, further, denied that there are planetarv
cosmological consequences that followed from the assumption that eccen­ eccentrics, epicycles, or elevations and depressions o f the planets. Should
trics and epicycles were real, material spheres. such things exist in the heavens, they would clearly imply that the fifth
body, or celestial ether, is divisible. For if a planet were elevated or depressed
— that is, varied its distance from the earth - it would follow that the
supposedly indivisible fifth element, or ether, would in fact be divided.
III. C o sm o lo g ic a l p ro b le m s w ith eccentrics and ep icycles
Under such circumstances, either the ether would fill the places left vacant
By enclosing each set o f eccentric planetary orbs within concave and convex by the planet as it moved higher or lower, or the places from whence the
surfaces that were themselves concentric with respect to the earth’s center, planet withdrew would remain void. Since neither o f these alternatives is
Ptolemy himself had seemingly made a strong gesture toward reconciling naturally possible, Avenalpetras rejected the existence o f eccentrics, epi­
his own cosmology with that o f Aristotle. In both systems, the earth’s cycles, and planetary variations in distance.5 *58
3
center was the center ot motion tor each total planetary orb. Natural phi­ Some fifteen or twenty years later, many o f the cosmological objections
losophers could thus continue to believe that the fundamental structure o f that were raised against the three-orb system o f material eccentrics and
Aristotle’s system was preserved: the external surfaces o f every planetary epicycles also appear, though not always clearly expressed, in Roger Bacon's
sphere were concentric with the earth. Ptolemy sharply diverged from treatises o f the 1260s. Perhaps the best statement o f them and their incom-
Aristotle, however, by his assumption that within the external concentric
54. For this approximate date, see William Wallace's article on Vincent in Dictionary of Scientific
surfaces that comprised each total planetary sphere or orb were three or
Biography. 1970-1980, 14:34-35.
more partial eccentric orbs with centers other than that o f the earth. In 55. “ Utrum aliquod spacium sit inter sphaeras planetarum.” Vincent o f Beauvais. Speculum
violation o f Aristotle’s dictum that all celestial spheres move with uniform naturale, bk. 3, ch. 104, 1624, ircols. 230-231.
56. “ Utrum inter sphaeras sit corpus aliquod vel spacium illud sit vacuum; an spnaerae
motion around the earth as center, Ptolemy assumed the motion o f all his
contingant se invicem." Ibid., col. 230.
eccentric spheres to be around points other than the earth's center. Although 57. For other instances o f Vincent’s mention o f “ Avenalpetras.” see ibid., cols. 226 (ch. 971
most scholastics recognized that eccentric orbs and epicvcles explained pla­ and 228 (ch. ioo). In a discussion o f Albertus Magnus’s Sumtna de creaturis, Gilson (1955.
281) identifies the name “ Anavelpetra” with the famous Arabian astronomer al-BitrujI
netary variations in distance and latitude that went unaccounted for in Ar­ (Gilson has “ al-Bitrogi” ). Is “ Anavelpetra" the same as “ Avenalpetras” ? But in his com­
istotle’s system ot concentric spheres, they were also aware o f a number ot mentary on De caelo, Albertus refers to al-Bitrujl as “ Alpetrauz” or “ Alpetragius.” but
significant problems in the received system o f eccentrics and epicycles that not Anavelpetra or Avenalpetras (see Albertus Magnus. De caelo, 1971, 276, under “ Al-
Bitrujl” ). Vincent, however, savs that Avenalpetras rejects eccentrics and epicycles, as
were potentially subversive of Aristotelian cosm ology and physics. As de­ indeed did al-Bitrujl (see BitrujI, 1952, 11). Thus the identification o f al-Bitrujl with
fenders o f the three-orb system, they sought to explain these anomalous Avenalpetras gains plausibility. In his Lucidator, differ. 6, 1988, 315, line 1, Peter o f Abano
refers to al-Bitrujl as “ Avempetras.”
53. Clavius. Sphere, ch. 4, 1593, 525; Amicus. De caelo, cract. 5. qu. 4, dubit. 3. art. 3, 1626, 58. “ Et dicit Avenalpetras quod ille sese contingunt. Sed ipse ponit quod non sunt eccentrici.
267, col. 1; Mastrius and Bellutus, De coelo, disp. 2, qu. 1. art. 2, 1727, 3:488. col. 2. nec epicycli, nec elevationes, nec depressiones planetarum, quod dicit ea ratione: quia
Aversa. De caelo, qu. 32. sec. 6—7. 1627, 66, col. 2—74. col. 1, seems to think o f his secundum naturam quintum corpus non est divisible. Si autem esset elevatio ac depressio
eccentrics and epicycles as cavities or channels in a firm, material celestial region. With planetarum tunc oporteret quod corpus illud divideretur ac succederent stellae elevatae
some reservation, we mav classify him as a qualified supporter o f eccentrics and epicycles. ac depressae in locum suum; aut quod locus unde recederet Stella vacuus remaneret.
Clavius’s lengthy discussion o f eccentrics and epicycles may be the most detailed and Quorum utrumque est impossibile secundum naturam. " Vincent o f Beauvais, Speculum
significant of all scholastic accounts (see Clavius. Sphere, ch. 4, 1593, 499—525). naturale, bk. 3, ch. 104, 1624, i:col. 230.
288 THE CELESTIAL REGION MOBILE CELESTIAL ORBS 289

patability with Aristotelian cosm ology was presented by Cecco d’Ascoli sibilities were ruled out by Aristotle and most o f his followers. A t the other
(1269-1327), who, in his De eccentricis et epicyclis, where he upheld the side o f the eccentric orb, the thinnest part will be unable to fill the space
existence o f eccentrics and epicycles, summarized the major impossible formerly occupied by the thickest part. In these circumstances, either a
consequences that opponents o f eccentrics and epicycles believed would vacuum will form, or, in order to prevent a vacuum, matter adjacent to
result from their existence. Cecco explains that “ if there were eccentrics orb c' must instantaneously fill any empty spaces. Under these circumstan­
and epicycles, then rarefaction or condensation would occur, which is im­ ces, either void spaces exist in the heavens, or celestial matter is divisible
possible by the first [book] o f [Aristotle’s De] celo et mundo; or a vacuum and capable o f rarefaction to fill a potential vacuum. In the Aristotelian
would occur, which is impossible, as is said in the fourth [book] o f [Ar­ physical world, neither alternative was acceptable.
istotle’s] Physics-, or there would be a separation o f the spheres, which is
impossible, as is obvious in the second [book] o f De celo et mundo; or there
would be a penetration o f bodies, which is false, as is obvious in the fourth 2. Are the celestial spheres continuous or contiguous?
[book] o f the Physics.” 59
The problems for Aristotelian cosm ology that Cecco d’Ascoli raised about
eccentrics were primarily about the relationships between the external sur­
1. Vacua and condensation and rarefaction in the heavens
faces o f any tw o successive orbs - that is, between the convex surface o f a
In Pierre d’A illy ’s language, these possibilities apply only to eccentric orbs contained sphere and the concave surface o f its containing sphere. In A r­
that are secundum quid, that is, eccentric orbs that have one eccentric surface istotelian terms, there are three possibilities: (1) the surfaces are continuous,
and one concentric surface. Bacon, d’A illy, and many others well into the that is, they coincide; (2) the surfaces are contiguous, that is, they are distinct
seventeenth century considered such eccentric orbs to be o f unequal thick­ but in direct contact at every point; or (3) they are w holly or partially
ness because the points o f apogee and perigee were unequally distant from distinct and without contact. Aristotle distinguished tw o kinds o f contact.
the center o f the w orld.60 The absurdities described by Cecco d’Ascoli In one w ay, things that are in succession and touch are said to be “ contig­
derived from the rotations o f such orbs. For example, if w e assume that uous.” Thus each o f tw o distinct surfaces in contact at all points would be
eccentric orb c' (see Figure 8) has rotated 180 degrees, it follows that the successive and contiguous and have the same shape.61 But if those two
thickest and thinnest parts o f it w ill have exchanged places. Because the surfaces became one and the same, they would be considered “ continuous.”
thickest part o f c' will occupy more space than the thinnest part, it must Indeed, “ continuity would be impossible if these extremities are tw o .” T o
make a space for itself by pushing away some o f the surrounding matter emphasize his point, Aristotle further explained that “ i f there is continuity
that now occupies the place it must enter; or it must occupy the same place there is necessarily contact, but if there is contact, that alone does not imply
with that matter. If it displaces the matter presently in that place, the dis­ continuity; for the extremities o f things may be together without necessarily
placed matter, in turn, must find a place for itself and therefore must divide being one.” 62 Aristotle used the concept o f continuity in his definition o f
adjacent celestial matter. Within the set o f planetary orbs that contains orb the “ place” o f a thing, which he defined as “ the boundary [or inner surface]
c', matter must condense somewhere, so that the thicker part o f c' can o f the containing body at which it is in contact with the contained body.”
occupy a greater place. T o accommodate the rotations o f such orbs, celestial Instead o f tw o surfaces, however, the surface o f the container and the surface
matter must be conceived as divisible and condensible, both o f which pos­ o f the thing contained formed only one surface and were therefore contin­
uous. O r, as Aristotle expressed it: “ place is coincident with the thing, for
59. “ Si esset ponere excentricos et epiridos, tunc m odo esset rarefactio aut condensatio, quod
boundaries are coincident with the bounded.” 63
est impossibile, ut patet prim o C e li et mundi; aut vacuum , et hoc est impossibile, ut dicitur
4° Phisicorum ; aut scissio sperarum, quod est impossibile, ut patet 2° C e li et mundi; aut Perhaps the first one to apply the concept o f contiguity to celestial orbs
corporum penetratio, quod est falsum, ut patet 40 Phisicorum.” C ecco d’ A scoli, D e eccen­ was Ptolemy, who, after presenting the distances from the earth o f the
tricis [Boffito], 1906-1907, 161. In her edition o f Peter o f A b a n o ’s Lucidator dubitabilium
successive planetary spheres in his Hypotheses o f the Planets, declares that
astronomiae, Graziella Federid V escovin i has reedited this w o rk o f C e c c o ’s as Appendix
II (Peter o f Abano, Lucidator, 1988, 383-394). B oth editions were made from the same
manuscript and are virtually identical. C e c c o ’s Latin text, as d ted in this note, appears
on page 384, lines 19-24. R oger Bacon also insisted that “ it is impossible to assume an 61. In D e caelo 2-4 .287a.6-7, Aristotle [Guthrie], i960, declares that “ w hat is contiguous to
eccentric orb o f any planet, because then it w o u ld be necessary that the celestial bo dy be the spherical is spherical.”
divisible; or that tw o bodies be in the same place; or that a vacuum exist.” Bacon, Opus 62. Aristotle, Physics 5 .3 .2272 .9 -12, 2 1 -2 3 [Hardie and Gaye], 1984.
tertium, 1909, 119; Bacon, D e celestihus, pt. 5, ch. 13, Opera, fasc. 4, 1913, 433-434. 63. Ibid., 4 .4 .2 12a. 5 -6 and 212a. 30. A ristotle’s doctrine o f place was as applicable to nested
60. For exam ple, in Figure 8, a' is thickest between points F B , where F is the point o f perigee, celestial spheres as it was to terrestrial objects, as is evident from his denial o f the existence
and thinnest at point A , the point o f apogee; whereas eccentric sphere c' is thickest at o f places beyond the w orld (D e caelo 1.9.2792.8-15), thereby im p lyin g the existence o f
OH, w here H is the point o f apogee, and thinnest at point K, the point o f perigee. places everyw here w ithin the world.
290 THE CELESTIAL REGIO N MOBILE CELESTIAL ORBS 291

If (the universe is constructed) according to our description o f it, there is no space convex surface o f one planetary sphere was exactly equal to the distance o f
between the greatest and least distances (of adjacent spheres), anci the sizes o f the the concave surface o f the next-highest celestial sphere. As Campanus ex­
surfaces that separate one sphere from another do not differ from the amounts we pressed it, “ the highest point o f the lower [sphere] coincides with the lowest
mentioned. This arrangement is most plausible, for it is not conceivable that there point o f the higher.’’6' In this brief passage, did Campanus consider the
be in Nature a vacuum, or any meaningless and useless thing. The distances o f the relevant surfaces continuous or contiguous? We cannot say with any cer­
spheres that we have mentioned are in agreement with our hypotheses. But if there tainty, since either alternative is compatible with the equality o f the distances
is space or emptiness between the (spheres), then it is clear that the distances cannot o f those surfaces. Also compatible with continuity or contiguity is Cam -
be smaller, at any rate, than those mentioned /'4 panus’s conviction that only by a fusion o f these two celestial surfaces could
waste space be avoided, either in the form o f a vacuum or as some kind o f
Campanus o f Novara repeated the substance o f Ptolem y’s position, as separate matter distinct from the orbs themselves. The Aristotelian defi­
we shall sec in the next paragraph. But when justification o f contiguity or nitions o f contiguity and continuity are compatible with the virtual or actual
continuity was required, scholastics turned to Aristotle. Christopher Clavius fusion, respectively, o f two successive surfaces, so that not only are their
reveals how Aristotle’s definitions o f contiguity and continuity were applied distances equal but nothing can lie between them. Without discussion o f
to celestial orbs. If a line were drawn from the center o f the world and the alternatives, or even any realization o f the issues involved or any aware­
intersected with the ninth and tenth heavens or spheres, Clavius argues that ness o f the choice they had made, most scholastic authors unknowingly
the point on the ninth sphere and its immediate neighbor on the tenth sphere opted for one or the other alternative, confident that they had neither fallen
would be two distinct points in the mind but one and the same point in into difficulty about planetary distances nor allowed matter or void to
actuality. Despite the oneness, and therefore the continuity, o f the successive intervene between successive spherical surfaces. O nly a few, including C lav­
points on the surfaces o f the tw o successive orbs, Clavius was aware that ius, were sufficiently knowledgeable to articulate the issues.
the ninth and tenth spheres had different motions and therefore insists that Even if the distinction between continuity and contiguity o f successive
the convex surface o f the ninth orb and the concave surface o f the tenth celestial surfaces could be ignored with respect to planetary distances and
orb were contiguous rather than continuous. Indeed, Clavius insists not intervening matter, it was o f crucial importance with regard to the diversity
only that the successive and immediate surfaces are contiguous but that o f celestial motions, as Roger Bacon recognized when he declared that
nothing can lie between them, tor otherwise an infinite process would result. continuous surfaces would cause those orbs to “ be moved with equal ve­
For example, if a globe were assumed in air, nothing could lie between the locity, even with the same motions, which is contrary to experience.’’66
convex surface o f the globe and the concave surface o f the air surrounding Although Bacon’s contemporary, Robertus Anglicus. thought he could
the globe. For if something, say body a, could intervene between the two reconcile continuity o f celestial surfaces with diversity of celestial m otions,67
surfaces, then something else would have to intervene between the convex Bacon, and probably most other natural philosophers and astronomers,
surface o f the air and the concave surface o f body a, as well as between the including Richard o f Middleton, Nicole Oresme, Albert o f Saxony, John
convex surface o f body a and the concave surface o f the air; and so on ad xMajor, and Christopher Clavius, believed that the obvious facts o f astron­
infinitum. In order to allow for distinct and even contrary motions and for om y required a denial o f continuity and the assumption o f contiguity.6S The
the motion o f a superior orb to be communicated to an inferior orb, Clavius
settled for successive celestial surfaces that were one and the same in reality 65. “ Per hoc enim sequitur quod supremum interioris sic intimum superions sue." Campanus
but conceptually distinct. Moreover, he chose to characterize them as con­ o f Novara, Theorica planetarum, 1971. 331; see also 331-337 and 53-55. From the Latin
text, we observe that Campanus uses no term for “ coincides." It was usually assumed
tiguous without the possibility o f intervening matter (Clavius [Sphere, ch.
in medieval Islamic and Latin astronomy that the distance from the earth o f the convex
i], Opera, 1611, 42). surface o f one planetary sphere was equal to the distance from the earth o f the concave
Clavius has made an important point, one that was probably implied in surface o f the next sphere (see the tables of distances and dimensions in ibid., 356—363).
In effect, since the two distances were identical, so were the surfaces. xMthough the
medieval discussions: if the convex surface o f an orb is continuous with or
distances o f the innermost and outermost circular surfaces were fixed, the distances of
contiguous to the concave surface o f the next successive superior orb, those the planets varied within their respective epicycles.
two surfaces will be equidistant from the center o f the earth. Whether 66. Bacon. Opus tertium, 1909, 123; Bacon, De eelestihus, pt. 5, ch. 13, Opera, fasc. 4, 1913,
successive spherical surfaces were continuous or contiguous was thus o f no 436 .
67 In his commentary on the Sphere of Sacrobosco, written around 1271. Robertus avoided
consequence with regard to the measurement o f planetary distances. In his the major problem with continuous orbs by assuming that the outer surfaces o f celestial
widely used Theorica planetarum, Campanus o f Novara declared that the6 4 orbs were immobile, with only their middle parts, which he likened to a fluid, being
capable o f motion. Under these conditions, each orb could move independently o f the
others. See Sacrobosco, Sphere, 1949. [4“ (Latin) and 202-203 (English).
64. From Ptolemy [Goldstein;. 1967. 8, col. 1. See also Ptolemy [Toomer], 1984, 4.40, n. 4- 68. Richard o f Middleton [Sentences, bk. 2, dist. 14, art. ), qu. 1], 1591, 2:184, col. 1; Oresme.
292 THE CELESTIAL REGION MOBILE CELESTIAL ORBS 293

different directions in which planets were carried by their orbs or the diverse impossibilities were it not for an otherwise unexplained synchronization o f
speeds at which they rotated made it obvious that successive celestial surfaces motions. O n the assumption that eccentric orbs move uniformly, d’Ailly,
could not form a single unified, continuous surface but had to be distinct for example, held that when the thickest part o f one eccentric is moved
entities, in contact at every point: that is, contiguous. Although Campanus toward its opposite side, another eccentric orb moves uniformly in the
o f Novara failed to express the distinction between continuity and conti­ opposite direction. When the tw o eccentric orbs, say a and b, have simul­
guity, the diversely directed motions o f the spheres makes it almost man­ taneously moved 180 degrees, the thickest part o f orb a will have come to
datory to attribute to him an assumption o f contiguity.69 For around five occupy the place formerly occupied by the thickest part o f orb b\ and
centuries, scholastics who assumed the existence o f celestial orbs probably similarly, the thinnest part o f orb a will also have come to occupy the place
assumed that the surfaces o f successive orbs were contiguous rather than formerly occupied by the thinnest part o f orb b . 70 In this manner a balance
continuous. is always maintained, and the dreaded impossibilities are perpetually
There was thus little reason to believe that extraneous matter or the avoided. H ow and w hy such synchronization o f orbs should occur is no­
dreaded void could intervene between successive surfaces o f celestial orbs where explained, but the idea was already presented early in the fourteenth
solely because they were contiguous or continuous. A more serious source century by Cecco d’Ascoli and repeated long after d’A illy by Georg Peur-
o f fear that intervening matter or void might intrude between surfaces arose bach in the fifteenth century and Christopher Clavius in the sixteenth and
for a quite different reason, namely from the idea that an eccentric material seventeenth centuries.71
orb that possessed Tane concentric surface and one eccentric surface was not
only o f unequal thickness but was ovoid in shape. If so, then Aristotle may 3. The rejection o f continuity and contiguity: the assumption o f
have furnished the basis for this mistaken notion when, in demonstrating matter between two orbs
the sphericity o f the heavens, he declared that if the w orld were not spherical
but “ lentiform, or oviform, in every case we should have to admit space The existence o f material eccentrics was made to seem viable in yet another
and void outside the m oving body, because the whole body would not way: b y the assumption o f intervening matter between the orbs, which
always occupy the same room ” (Aristotle, Decaelo 2.4.287a. 12-24 [Stocks], implied the rejection o f spheres that were either continuous or contiguous.
1984). In a brief though important passage, Vincent o f Beauvais describes such an
In commenting on this passage, N icole Oresme distinguished different opinion when he reports:
circumstances under which oval-shaped planetary orbs would or would not
produce the impossible consequences described earlier. He argues ([Le Livre Some say that the spheres do not mutually touch and that a body of the same nature
du del, bk. 2, ch. 10], 1968, 391) that “ if the planetary spheres were oval lies between them. Indeed [that body] is divisible by the spheres but is not trans-
in shape, being moved in a manner different from the sovereign [or last] mutable into another species, and in it eccentrics and planets are elevated and de­
heaven and on different axes. . . , either there w ould have to be an empty pressed [i.e., vary their distances from us]. Nevertheless, some ancients say that
place or penetration in the heavens — that is, one heaven w ould pierce
70. See d ’A illy, 14 Questions, qu. 13, 1531, 163V, for the argument and 164V for d ’A illy ’s
through the other —or there would have to be condensation or compression, brief response.
all o f which are impossible in nature.” 71. B etw een 1322 and 1327, C e c c o d ’Ascoli, D e eccentricis [Boffito], 1906, 166, described the
same mechanism for synchronizing the motions o f eccentric orbs b y what he called
Judging from certain responses, some scholastics seem to have believed
“ proportional m otions” (proportionates motus). A lth o u gh C ecco, w h o defended the exis­
that eccentric orbs o f uneven thickness would indeed produce the alleged tence o f eccentrics and epicycles, thought the idea o f “ proportional m otions” was a good
idea, he denied that P tolem y had it in mind. Peurbach, Theoricae, 1987, 10, used the same
expression, “ proportional m otions,” when he declared that “ the deferent orbs o f the
Le Livre du del, bk. 2, ch. 9, 1968, 385; A lbert o f Saxony, D e celo, bk. 1, qu. 4, 1518,
apogee o f the Sun m ove b y their ow n proportional m otions, so that the narrower part
89V, col. 1; and Major, Sentences, bk. 2, dist. 14, qu. 4], 1519b, 75V, col. 1. In his D e
o f the superior is always above the wider part o f the inferior, and go around equally
multiplicationes specierum. Bacon declares unequivocally that “ the spheres are contiguous
fast.” Som e tw o hundred years after d ’A illy, C lavius approvingly presented the same
and possess distinct surfaces” (Bacon [Lindberg], 1983, 119). In the late sixteenth century,
explanation. See C lavius, Sphere, ch. 4, 1593, 521 (for a description o f the impossibilities)
Clavius insisted that the celestial orbs are contiguous, w ith each superior orb including
and 523 (for the response). T h e idea m ay be traceable to Bernard o f Verdun, at the end
its imm ediately inferior orb “ and there is no m edium between one and the other, as in
o f the thirteenth century. A fter m entioning the usual charge that eccentrics w ould produce
the peels o f an onion, where everyw here w e see the upper [peel] surround a low er [peel]”
the dreaded impossibilities, Bernard explains that the “ different parts” o f eccentric orbs
(Sunt autem omnes orbes coelestes contigui prorsus et im m ediati inter se, ita ut semper
“ succeed themselves continually in the points or places o f the farther and nearer distance”
superior inferiorem indudat. nihilque inter unum atque alterum sit m edium non secus ac
- that is, in the points o f apogee and perigee - “ that are im agined in the con vexity o f the
in tunicis caeparum videm us superiorem undique circundare inferiorem). Clavius, Sphere,
surrounding orb.” T h e translation is mine from Grant, 1974. 523 (I have here replaced
ch. 1, Opera, 1611, 3:10.
“ longitude” w ith “ distance” ). Bernard did not use an expression comparable to “ pro­
69. As do Benjamin and T oo m er (see Cam panus o f N ovara, Theorica planetarum, 1971, 412,
portional m otion(s).”
n. 47), and Aiton, 1981, 90.
294 THE CELESTIAL REGION MOBILE CELESTIAL ORBS 295
eccentrics and planets traverse the b o d y that lies betw een the spheres but yet do not “ Because o f this,” Albertus concludes, “ I say that they [the successive
divide it. T his occurs because ot the form ality o f these bodies, ju s t as light [lumen] spheres] never touch but that intervals [or gaps between the spheres] in
traverses through the air [and does not divide it].7' some particular place are sometimes greater and sometimes smaller, and
that a rare or dense body existing between the circles [or spheres] fills them .”
Whereas Vincent o f Beauvais was content to report the opinions o f others Albertus adopts this interpretation and identifies it with “ the opinion o f
using the anonymous phrase “ some say,’’ Albertus Magnus not only pro­ Thebit, a wise philosopher, in a book which he composed on the motion
vided a much clearer and more complete account in his commentary on De o f the spheres.” 77 Thus every sphere is separated from its immediate neigh­
caelo,7
273 but he unhesitatingly adopted the concept o f a divisible matter in­ bors by a certain kind o f celestial matter that is capable o f rarefaction and
tervening between the planetary orbs. Like Bacon, Albertus seems to have condensation.
considered eccentrics as ovoid in shape and was therefore convinced that if That Albertus adopted such an opinion - and it seems that he did - is
eccentrics are nested one within the other their motions would cause gaps quite astonishing. It marked a radical departure from Aristotle’s cosmology.
between their surfaces.74 Because Albertus believed that these gaps or spaces Indeed, Albertus distinguished the matter that lies between orbs from the
could not be void, one ot two alternatives must occur: either the various ether, or fifth element, that composes the rest o f the celestial region. Unlike
eccentric spheres would rarely and condense, to prevent formation o f a ether, which is incorruptible, indivisible, and therefore suffers no rarefaction
vacuum; or another body must intervene between any two successive ec­ or condensation, the intervening matter can rarefy and condense and must
centric spheres. therefore be divisible. Although the creator o f the celestial orbs made some
To refute the first alternative, Albertus argues that if the circles or spheres parts permanently rarer and some parts permanently denser than other
rarefied and condensed, their shapes would vary. Hence their motions parts,78 he made “ the intervening body contractable and extendable so that
would be essentially unknowable, and the data derived from those motions 77. "Et haec est sententia Thebit philosophi sapientis in libro, quern composuit de motu
would be false.75 If, however, we say that the circles, that is, spheres, sphaerarum.” Ibid. Hossfeld, the editor o f Albertus’s De caelo, identifies his source as
themselves are sometimes in direct contact because no medium intervenes Thabit’s (or Thebit’s) De motu octavae sphaerae (see Hossfeid’s note to line 29 on page 30).
In Carmody’s edition o f this treatise (see Thabit ibn Qurra [Carmody], i960, 102-107),
between their surfaces, then, when the surfaces o f these eccentric orbs do there is no such passage. Hossfeid’s claim, however, is based on a single manuscript,
separate because o f their shapes, a new body that did not exist there before Paris, Bibliotheque Nationale, t'onds latin, MS. 7195, I40vb-i43va, which is but one o f
cannot be generated to fill the void which must inevitably occur. That no a number o f manuscripts that Carmody used for his edition. If such a passage existed in
Thabit’s De motu octavae spherae, it is unlikely that all traces o f it would have disappeared
other body could have existed there before the separation o f the surfaces is from Carm ody’s edition. Indeed, the theory o f intervening matter between the celestial
obvious, because if such a body did exist there before separation, then two orbs does not appear in any other o f Thabit’s works that were translated into Latin and
bodies would have occupied the same place, which is impossible."6 which have been edited by Carmody.
However, according to Bernard Goldstein (1986, 277-278), Albertus, in an unpublished
part o f the fourth book o f his Sentences, cited "a passage from Thabit’s lost treatise. Libro
72. “ Quidam etiam dicunt quod sphaerae non contingunt se et corpus eiusdem naturae est de exccntricitate orbium," in which "w e are told that there is a subtle matter that fills the
inter eas. Divisible quidem a sphaeris sed non transmutabile in speciem aliam et in illo space between the spheres. This matter is uniform, transparent, and subject to division
elevantur et deprimuntur eccentrici et planetae. Quidam tamen antiqui dixerunt quod [i.e., fluid], but not to alteration.” Moses Maimonides also describes a similar interpre­
eccentrici et stellae transeunt per corpus quod est inter sphaeras et tamen ipsum non tation and attributes it to Thabit (Guide, pt. 2, ch. 24, 1963, 2:325). But it is probably
dividunt. Et hoc contingit propter formalitatem ipsorum corporum. sicut et lumen transit from Maimonides, rather than Thabit (B. Goldstein, ibid., 278), that Levi ben Gerson
per aerem.” Vincent ot' Beauvais, Speculum naturaie, bk. 3, ch. 104, 1624, i:col. 231. derived the idea that the celestial region "consists o f planetary shells separated by fluid
73. Albertus Magnus. De caelo, bk. t, tract. 1. ch. 11. Opera. 1971. vol. 5, pt. 1:29-30. lavers with certain properties that allow us to compute their thicknesses” (ibid., 273). In
Weisheipl (19K0. 27) believes that Albertus wrote all ot' his Aristotelian paraphrases, the Hebrew text o f his Astronomy, Levi sought to compute these thicknesses, but the
including that on De caeio, between 1250 and 1270. Thus his account may have been calculations were apparently not incorporated into the Latin translations o f his treatise
written after Vincent’s report. (ibid., 285). Precisely what Thabit may have had in mind is thus lelt vague. On Mai­
74. Because Albertus assigned only one eccentric to each pianet, we may infer that he was monides and Thabit, see also Duhem’s brief discussion, Le Systeme, 1913—1959, 2:118—
not reporting Bacon’s "modern” three-orb system. U 9-
75. “ Si enim nos diceremus, quod ;psi circuli ranficantur et inspissantur. tunc non semper The ultimate source ot the theory o f separate matter lying between successive orbs
tenerent figuram eandem, et sic non posset sciri motus eorum, quod constat omnibus may have been Ptolemy himself, when, after describing the contiguity and contact o f
illis esse falsum. qui sciunt canones motus excentricorum. ” Albertus Magnus, De caelo, successive surfaces o f orbs in his Hypotheses of the Planets, he allows that " if there is space
bk. 1, tract. 1, ch. 2, Opera, 1971, 5, pt. 1:30, col. 1. or emptiness between the (spheres), then it is clear that the distances [between successive
76. "Et si nos diceremus. quod aliquando ;ta se tangunt circuli, quod nihil est medium surfaces o f successive planetary orbs] cannot be smaller, at any rate, than those men­
ipsorum, tunc oporteret nos concedere, quod tunc intercideret vacuum inter cos. quando tioned.” See Ptolemy [Goldstein], 1967, 8, col. 1 (for the full passage, see Sec. III.2 o f
per motum distant a tali situ coniunctionis. quia quando seiunguntur, non generatur ibi this chapter).
novum corpus, quod prius non tuit. Prius autem nullum potuit esse medium, si omnino 78. See the distinction between static and dynamic rarity and density, in Chapter 10, Section
et in omni parte se tangendo impleverunt, quia si fuisset ibi tunc aliud corpus, fuissent II. 1. c. Albertus Magnus denies dynamic rarity in the regular celestial ether from which
duo corpora in eodem ioco, et hoc est impossibile.” Ibid. thbqarbs are composed but allows it for the special matter that lies between orbal surfaces.
296 THE CELESTIAL REGION
MOBILE CELESTIAL ORBS 297
it should fill what lies between the spheres.” 79 Thus did Albertus abandon
orbs, the external surfaces o f which are concentric.83 Here, o f course, was
the important Aristotelian concept o f celestial homogeneity and assume
the fundamental feature o f the compromise between Ptolemaic astronomy
instead the existence o f tw o different kinds o f eternal celestial substances:
and Aristotelian cosmology. With the extreme surfaces o f every planetary
one divisible, and therefore changeable; the other indivisible and unchange­
sphere assumed concentric with respect to the earth, the eccentric orbs
able. Moreover, in his interpretation, the celestial orbs were no longer in
contained within those concentric surfaces could possess their ow n centers
direct contact.
without any adverse effect on Aristotelian cosmology.
Few chose to follow Albertus’s radical theory.80Convinced that eccentrics
were essential to account for the astronomical phenomena, Albertus was
obviously prepared to abandon certain important Aristotelian concepts in 5. Eccentrics and the problem o f a plurality o f centers
favor o f a system that would save the phenomena and also preserve a viable
But even i f the earth could serve as a center o f the universe, the existence
cosm ology. As we shall see, others proceeded in a similar fashion.
o f at least one other center for the eccentric orbs would involve at least tw o
different centers for celestial bodies. If tw o such centers existed, could a
4. I f eccentrics exist, can the earth lie at the center o f the world? heavy body m ove naturally downward to its natural place, when the latter
is defined as a unique center, coincident with the earth’s center, that func­
Although the potential impossibilities just described were probably consid­
tions as a terminus ad quern? According to d’Ailly, some denied that a heavy
ered the most serious cosmological difficulties for scholastic authors, a num­
body could reach its natural place at the center o f the world, arguing that
ber o f other objections appeared rather regularly. D ’A illy reports that some
a heavy falling body would either have to m ove to both centers simulta­
questioned whether, if eccentrics existed, the earth could lie at the center
neously or, because it could not choose between them, would not move at
o f the universe.8' This objection was apparently based on the assumption
all.84 D ’A illy responded that despite the different centers, every heavy body
that all celestial orbs are eccentric. If so, the earth could not be their center,
w ould nonetheless move toward the center o f the world, because the latter
by definition. But d’A illy and others replied that the earth lies at the center
is the center o f the “ total orb,” that is, the center o f all the concentric
o f the “ total orb,” that is, it lies at the center o f all the concentric surfaces
surfaces which enclose all the eccentric orbs.85 In a similar vein, Johannes
that serve as boundaries for each set o f planetary orbs.82 As Albert o f Saxony
Versor argued that a plurality o f eccentric and concentric centers would not
explained, the absence o f the earth from the center o f eccentric orbs posed
render meaningless the idea o f a unique, absolute “ dow n” location. “ D o w n ”
no problem, because eccentrics are included within the totality o f planetary
in the universe, Versor explains, was usually taken “ in relation to the whole
heaven, or in relation to a w hole orb, but not with respect to partial circles
79. Albertus M agnus, D e caelo, bk. 1, tract. 1, ch. 2, 1971, 5, pt. 1:30. Albertus adds that
this is also the opinion o f Avicenna (in the D e caelo et mundo o f the latter’s S u fficie n t)
[or orbs].” 86 But the heaven, “ is concentric with respect to each o f its
and Averroes (in the D e substantia orbis, w hich Albertus cites as Liber de essentia orbis). extremal surfaces; and the same holds for any orb, even though a partial
80. O n e w h o did was C ecco d’Ascoli, w h o declared that “ orbs are neither continuous nor
contiguous, but there is an intervening b o d y between them, w hich, according to Thebit
83. A lbert o f Saxony, D e celo, bk. 2, qu. 7, 1518, io6r (mistakenly foliated 107), col. 2, for
and Albertus, is capable o f being com pressed.” For the Latin text, see C e c c o d ’Ascoli,
the objection; io 6v, col. 2, for the reply.
Sphere, 1949, 353. In the seventeenth century, G iovanni Baptista Riccioli cited this very
84. D ’A illy, 14 Questions, qu. 13, 1531, 163V, w ho reports this argument, speaks o f on ly tw o
passage as evidence that C ecco believed in the fluidity o f the heavens (see Riccioli, Al~
centers, one for the w orld and the other for all eccentric orbs. B u t eccentric orbs had
magestum novum, pars post., bk. 9, sec. 1, ch. 7, 1651, 239, col. 2). In his D e eccentricis et
m any centers, because differences in their planetary eccentricities precluded a com m on
epicyclis, C ecco makes no mention o f bodies intervening between successive orbs. A l­
center. Paul o f Venice, Liberceli, 1476, 31, also spoke o f tw o centers, but A lbert o f Saxon y
though he seems not to have adopted it, T h om as Aquinas mentions the theory without
spoke o f “ m any centers” (plura centra) (De celo, bk. 2, qu. 7, 1518, io6r, col. 2, for the
reference to Albertus when he explains (in his com m entary on Boethius’s D e trinitate)
objection; io 6v, col. 2, for A lb ert’s reply).
that supporters o f eccentrics and epicycles believe that this opinion avoids the dilemma
85. D ’ A illy, ibid., 164V. Paul o f Venice offered the same solution (Liber celi, 1476, 31),
that tw o bodies m ight have to occupy the same place and that the substance o f the spheres
explaining that the earth is “ in the m iddle o f the total orb o f the planets, because the orb
could be divided. Thom as describes the intervening matter as “ another substance, which
is totally concentric to the w orld ” (tamen est [i.e., the earth] in m edio totalis orbis
lies between the spheres and w hich, like air, is divisible and w ithout thickness, although
planetarum eo quod orbis totaliter est concentricus m undo). A lb ert o f Saxony, D e celo,
[unlike air] it is incorruptible.” For the Latin text, see Litt, 1963, 348. Robertus Anglicus
bk. 2, qu. 7, 1518, io6r, col. 2, presented the same objection w ith much the same response
mentions “ another opinion” (alia tamen opinio) in w hich matter is assumed between
orbs. See Robertus Anglicus, Sphere, lec. 1, 1949, 14 7-14 8 (Latin); 203 (English). on io 6v, col. 2.
86. Versor [D e celo, bk. 2, qu. 9: “ W hether eccentric, concentric, and epicyclic circles (i.e.,
81. D ’A illy, 14 Questions, qu. 13, 1531, 163V.
orbs) are to be assumed in the heavens to save the appearances o f the planetary m otions” ],
82. Ibid., 164V. Paul o f Venice accepted the same argument and also used the expression
1493, 22v, col. 1, for the objection, and 23r, col. 1, for V ersor’s reply. Here w e see the
“ total orb o f the planets” (totalis orbis planetarum); see Paul o f Venice, Liber celi, 1476, 3
com m on distinction that m ost natural philosophers drew between “ the w hole orb” (orbis
col. 2 (the last tw o lines; because the w o rk is unfoliated and is provided w ith few sig­
totalis, or as Versor put it, orbis integer), which embraces three or more eccentric orbs,
natures, the page numbers have been determined b y counting from the beginning o f the
Liber celi et mundi). and a “ partial circle” or “ orb” (circulus [or orbis) partialis), w hich refers to o n ly one o f the
constituent eccentric orbs o f a “ w hole orb.”
M O BILE CELESTIAL ORBS 299
298 THE CELESTIAL REGION
earth in this manner, it could do so only by a rectilinear motion or by a
surface o f one part o f the orb has a center distinct from the center o f the
w orld.” motion compounded o f rectilinear and circular motion. T o move toward
or away from the earth rectilinearly, a celestial body would have to be either
Although the concentric surfaces at each planetary sphere enabled the
heavy or light, or compounded o f both; or it might have an entirely different
earth to retain its cosmic centrality, and although the earth remained the
nature. But rectilinear motion toward or away from the earth and out o f a
natural place o f heavy bodies, the defenders o f solid eccentrics and epicycles
circular orbit would involve a celestial body in violent action, which was
had made a significant departure from Aristotelian cosmology: they allowed
contrary to the nature o f the celestial ether.89
celestial bodies to move around more than one center. Eccentric orbs were
The usual response was to deny that variations in planetary distances were
assumed to move around their own centers rather than around the earth as
the result o f rectilinear motion. D ’A illy argued90 that upward and downward
center. To accept the three-orb system as truly representative o f the physical
motion could happen only where generation and corruption occurred,
cosmos was to admit that, contrary to Aristotle, Averroes, and Maimonides,
namely in the terrestrial region.91 Paul o f Venice met the same objection
planetary spheres could rotate around geometric points other than the center
by a different argument (Liber cdi, 1476, 3 1, col. 2). To qualify as rectilinear
o f the earth, that is, other than the geometric center o f the universe. Most
ascent and descent, motions must be measured along a radius o f the world.
scholastics passed over this significant shift with little or no comment, but
Such measurements were therefore not applicable to circular motion, from
a few, like Cecco d'Ascoli, Nicole Oresme, and Jean Buridan met the issue
which it followed that the motion o f planets on circular eccentrics and
head-on. Because he believed that the celestial orbs were not all o f the same
nature and that celestial bodies differed in matter and form and in their epicycles did not qualify as rectilinear.
motions, Cecco insisted: “ it is therefore not absurd that they [the planets]
should have different and immobile centers.” 878 Oresme flatly declared:
7. The problem with epicycles
“ whether Averroes likes it or not, w e must admit that they [the heavenly
bodies] move around various centers, as stated many times before; and this It would appear that the acceptance o f eccentric orbs also implied a com­
is the truth” (\Le Livrc du del, bk. 2, ch. 16], 1968, 463). Buridan firmly mitment to epicyclic orbs. But at least one scholastic, Jean Buridan, accepted
stated: “ The Commentator [Averroes] speaks improperly when he says that the former but not the latter. Epicycles posed a special problem, because
the spheres are located by a [common] center; . . . modern astronomers [as- o f the M oon’s observed behavior. Since the Moon always shows the same
trologi] do not concede that all celestial spheres have the same center; indeed, face to us, Aristotle had argued that it cannot be said to rotate or revolve. /
they assume eccentrics and epicycles” ([De caelo, bk. 2, qu. 14], 1942, 191, O n the assumption that all planets are alike in their basic properties, he
lines 19-23). inferred (De caelo 2.8.290a.25-27) from the M oon’s behavior that no planets
rotated around their own axes. Aristotle’s denial o f rotation to the Moon
and ether planets played a significant role in arguments about the reality o f
6. Would planets move with rectilinear motion i f eccentrics
and epicycles existed? material epicycles.
Although the fundamental problem about epicycles is traceable to Roger
Because a key purpose o f eccentrics and epicycles was to account for changes Bacon in the thirteenth century,92 it was Jean Buridan and Albert o f Saxony-
in planetary distances from the earth, it was alleged that eccentrics and in the fourteenth century who described the two approaches available to
epicycles would cause planets to ascend and descend rectilinearlv as they natural philosophers.91 Buridan discussed the issue in a question on
alternately approached and withdrew from the earth. Following Albumasar,
Bacon held that motion is threefold: namely, from the center o f the world
89. “ But there is no violence in the heavens, as Aristotle says in the book On the Heaven and
(media)- toward the center; and around the center. '* Celestial bodies move the World ami in the eighth [book] ot'the Physios-, and it is obvious that nothing perpetual
only around the center o f the world, that is around the earth. For if a planet is violent.” Bacon, ibid., 444, lines 26-29.
moved around another center, it would sometimes be nearer the earth and 90. D ’Aillv, 14 Questions, qu. 13, 1531, 163V.
91. Ibid., 164V. Albert o f Saxony had earlier presented the same objection and resolution (see
sometimes farther away. But if a celestial body varied its distance from the his De celo, bk. 2, qu. 7, r5 r8, io6r. col. 2, for the objection, and io6v. col. 2, for the
87. In replying to an argument against eccentrics. Cecco declares: “ Si dicatur omnes orbes response).
92. Bacon, Opus tertium, 1909, 130—131; the De celestibus omits this section. Duhem, Le
esse eiusdem nature, quod est t'alsum, ut diat Albertus in libro Cdi ft numdi, et quia
Systeme, 1913-1959, 3:436, conjectures that Bacon may have been the first to propose
corpora celestia diversa sunt in torma et materia, et in mom diversa erunt: non ergo erit
this objection to the existence o f solid epicycles. Since Bacon speaks as if others had
inconveniens quod habeant diversa centra et inmobilia." Cecco d’Ascoli. De eccentricis,
already proposed the criticism, this seems unlikely.
1906, 167; see also the edition in Peter o f Abano. Lucidator, 1988, 393.
93. See also Chapter 17. Section IV.3a.1ii. for a further discussion o f these ideas in connection
88. Bacon, De celesttbus, pt. 5, ch. 17, Opera, fasc. 4, 1913. 444, lines 10-11, cites Albumasar’s
De conjitncaonibus as his source. with the Moon and its spots.
300 THE CELESTIAL REGION MOBILE CELESTIAL ORBS 301
“ Whether epicycles are to be assumed in celestial bodies.” 94 He based his Buridan then derives the following consequence: if the M oon does not
opinion on the behavior o f the “ man in the M oon ,” that is, the spot on have a proper motion around its ow n center, it cannot have an epicycle.
the lunar surface that had the appearance o f a man whose feet always point For i f the M oon had an epicycle but lacked a proper motion, the head and
toward - or lie at - the bottom o f the Moon. Buridan argued that if the feet o f the man in the M oon would change positions every time the epicycle’s
Moon had an epicycle, the man’s feet should sometimes appear in, or point apogee and perigee rotated 180 degrees. Because no such change is observed,
toward, the upper part o f the lunar disk. Thus if the man’s feet are at the Buridan concludes that the M oon can have no epicycle, from which he
bottom o f the lunar disk when the M oon is in the aux, or apogee, o f the generalizes that “ if an epicycle is not posited in the orb o f the Moon, it
epicycle, the feet ought to be in the upper part o f the lunar disk when the ought not to be posited in the orb o f the other planets, since all the reasons
Moon reaches the opposite o f the aux, or perigee, o f the epicycle. But such which apply to the other planets should also apply to the M oon” (Grant,
an occurrence is never observed. The feet always remain at the bottom o f 1974, 525). Thus did Buridan conclude that “ all appearances can be saved
the lunar disk, thus calling for the rejection o f an epicycle.95 Buridan suggests by eccentrics [alone] without epicycles” (ibid., 526).
a way to account for this phenomenon and retain the epicycle. We would O ne response to Buridan was to allow that a particular planet might
have to assume that “just as this epicycle is moved around its proper center, indeed behave differently from its sister planets. Albert o f Saxony adopted
so also is the body o f the M oon m oved around its proper center in a motion just such a strategy. After describing the problem much as Buridan had,97
contrary to that o f the epicycle and with an equal speed” (Grant, 1974, 526). Albert assumes that the M oon possesses a proper motion around its own
O nly in this way will the upper part o f the man always appear in the upper center in a direction that is contrary to the motion o f its epicycle. As for
part o f the lunar disk.
those w ho say that “ other planets do not have proper morions around their
Assuming with Aristotle that all planets possess the same fundamental proper centers, therefore the M oon does not,” Albert counters, without
properties, Buridan infers that i f the M oon has a proper rotatory motion, elaboration and perhaps with Buridan in mind, that the M oon’s nature
all the other planets should also possess that same motion. In agreement differs from that o f the other planets because the M oon’s upper and lower
with Aristotle, however, he was convinced that no planet could rotate parts can affect sublunar things differentially. Its proper motion around its
around its own center. Planets not only m ove from one position to another; ow n center is, therefore, not superfluous but brings the lower part o f the
they also cause transmutations in sublunar bodies. Consequently, i f planets M oon to the upper part and the upper to the lower. Because the M oon’s
rotated around their own centers, the rotations ought to affect the way in proper motion is contrary to the motion o f its epicycle, we do not observe
which they cause sublunar effects. That is, each planet ought to produce these continuous and regular turnings o f the spot in the M oon.98
differential effects; otherwise its rotatory motion w ould be superfluous. A lbert’s interpretation prevailed and was repeated with the same argu­
Taking the Sun as exemplar, Buridan argues that it does not produce such ments by Christopher Clavius in the numerous editions o f his commentary
differential effects, probably because it is a uniform, homogeneous body on the Sphere o f Sacrobosco that appeared in the late sixteenth and early
whose upper and lower parts are identical. A n y rotatory motion by the Sun seventeenth century.99 Although both Albert and Buridan sought to save
around its own center would therefore be superfluous, because no sublunar the observed behavior o f the spot in the M oon, they did so in radically
changes would result. “ But i f the Sun does not have such a motion, it does different ways. Whereas Buridan insisted on the uniformity o f planetary
not seem reasonable that the M oon should have it, since the Sun is much behavior and properties, Albert permitted divergence. Buridan sought for
nobler than the M oon.” 96 consistency: either all planets rotated around proper centers, or none did;
94. Buridan [Metaphysics, bk. 12, qu. 10], 1518, 73 r- 73V. T h e quotations are from m y either all planets moved on epicycles, or none did. The astronomical ap-
translation o f this question in Grant, 1974, 524-526.
95. Buridan is only partially correct. I f the M o o n were carried on an epicycle but lacked 97. A lbert o f Saxony, D e celo, bk. 2, qu. 7, 1518, io6r, col. 2 (the fifth principal argument).
rotatory motion, the appearance o f the M o o n w ould indeed change. B u t the change 98. A lth o u gh A lb ert o f S ax on y admitted that he had often seen a black spot in the M oon ,
w ould not be as Buridan describes it. T h e “ man in the M o o n ” w ould not turn upside
he denied that it resembled a man.
d ow n from apogee to perigee, but rather a terrestrial observer w ou ld see the man in the 99. In agreem ent w ith A lb ert’s position w ere d ’A illy (14 Questions, qu. 13, 1531, 163V, for
M o o n for awhile and then not see him. For this interpretation, I am indebted to m y the objection, and 164V for the response) and Paul o f Venice, Liber celi, 31, col. 2, for
student, M r. James Voelkel, and to an anonym ous reader.
the objection and 32, col. 1, for the response. Paul argued that because the M o o n has
96. “ E t si sol non habeat talem m otum nec videtur rationabile q uod luna habeat, cum sol sit “ diversity in its parts,” it requires a proper m otion, whereas the other planets lack diversity
m ulto nobilior quam luna.” Buridan, Metaphysics, bk. 12, qu. 10, 1518, 73V, col. 1. and need no proper m otions. W ithout in vokin g diversity, Bernard o f Verdun (Grant,
Buridan believed that planets were also unlikely to have proper m otions, because each
1974, 523-524) retained the lunar epicycle and also assumed that the M o o n som ehow
such m otion w ould require a special m over. W e w ould then have to assume “ as many
turns, or is turned, so that “ the spot always appears to us in the same shape [or form ].”
intelligences as there are stars in the sky, because each star w o u ld require a special mover A lth ou gh this interpretation was quite traditional b y the time C lavius wrote, he attributes
for its special m otion.” B ut “ Aristotle did not assign [or concede] such a multitude [of it to Jean Fem el (1497-1558). See Clavius, Sphere, ch. 4, 1593, 522, for the objection,
motions and intelligences].” Ibid. (Latin) and Grant, 1974, 526 (English).
and 525 for his response.
M O BILE CELESTIAL ORBS 303
302 THE CELESTIAL REGION
system o f concentric spheres as he described it in the twelfth book o f his
pearances could be saved only i f planetary homogeneity and uniformity
Metaphysics was inadequate. They sometimes reacted to that description in
were preserved. B y contrast, Albert o f Saxony thought it more important
strange ways. Thus in the seventeenth century, a Scotistic commentator,
to save the appearances than to preserve the uniformity o f planetarv be­
perhaps Hugo Cavellus (1571-1626), offered a lengthy analysis o f Aristotle’s
havior. In Albert’s scheme, it was not necessary that all planets should move
concentric spheres, explaining how Aristotle arrived at 55 and how one
on epicycles (the Sun did not). N or, as we saw, was it necessary that either
could reduce this to 47, which he mistakenly believed was Aristotle’s final
all planets or no planets move around their own centers. If the phenomena
total.103 O ur commentator declares that a single, uniformly moving, con­
could be saved by assuming that some planets really moved around their
centric sphere could not properly represent the motion o f a planet, because
own centers and others did not, Albert was satisfied.10010 2
planets alter their speeds and seem to change directions, m oving sometimes
In the seventeenth century, Melchior Cornaeus agreed with Buridan and
directly and sometimes retrogressively. T o take these anomalies into ac­
rejected the existence o f a lunar epicycle. As we shall see later (Ch. 14. Sec.
count, Ptolemy and other astronomers utilized deferents (presumably ec­
VIII. i.b.iii), however, the context o f his discussion and the reasons for his
centrics) and epicycles. But Aristotle also recognized that a single planet
decision were radically different.
had more than one motion and accounted for this by building on the systems
o f Eudoxus and Callippus: that is, he assigned a plurality o f concentric
8. Summary o f differences with Aristotle spheres to account for the motion ot each planet.104 In describing how
Aristotle did this, our commentator makes no further mention o f eccentrics
Although a few other arguments were sometimes cited for and against or epicycles. Indeed, it is as if he had equated the two systems simply
eccentric and epicvclic spheres,'01 those mentioned here were unquestion­ because both assigned a plurality o f spheres - not just one - to account for
ably the most important for cosm ology. Despite the widespread conviction the motion o f each planet. In light o f this, we are not surprised that, despite
that eccentrics and epicycles saved the astronomical phenomena and that writing in the seventeenth century, our commentator finds no reason to
Aristotelian concentric astronomy did not: scholastic natural philosophers mention Copernicus or the Copernican system. Aristotle’s system is ana­
were also aware that those same eccentrics and epicycles appeared to violate lyzed as if the Ptolemaic and Copernican systems had never existed.
important aspects ot Aristotelian cosm ology. In order to save the astro­
nomical phenomena and avoid alleged cosmological impossibilities, some,
and in a number ot instances many, natural philosophers made significant 9. On the physical nature o f eccentrics
departures from Aristotelian cosmological principles. Am ong the most sig­ Despite a rather large number o f authors who considered the suitability o f
nificant were the assumptions that (1) eccentric celestial orbs move with eccentrics for astronomy and cosmology, relatively few ventured opinions
circular motion around centers other than the earth; (2) that the M oon and about the nature o f such spheres, especially the deferent orbs. Were they
all other planets have proper motions around their own centers in a direction hollow and void, or filled with some rare or dense substance? O r were they
opposite to that o f their epicycles; (3) that successive orbs are not in direct solid? Were the spherical epicycles carried around as immobile bodies within
contact and the space between those orbs is occupied by a celestial substance their deferent orbs? O r did they move through the orb itself? Were the
that is divisible, though incorruptible; and, finally, (4) that celestial bodies, planets carried within an epicycle, or were they self-moved? As we shall
and therefore the celestial substance, need not be hom ogeneous.104 see, most o f those who did consider the internal nature o f the orbs did so
Scholastic commentators were aware that Aristotle’s description o f his in the late sixteenth and the seventeenth century. Few scholastics troubled
to describe the nature o f the spheres themselves, especially the deferent orb,
100. Nonetheless, one may ponder why Albert did not infer from the lunar rotation the which carried a spherical epicycle.
rotation o f all planets around their respective axes. Perhaps he thought, as did Paul ot
Venice later (see note 99 to this chapter), that they lacked the Moon's diversity and 103. See Cavellus, Metaphysics, bk. 12, summa 2, ch. 4, 1639, 4:448-450. Aristotle offered
therefore did not require axial rotation as they were carried by their respective epicycles. 49 as an alternative to 55.
101. For example. Bacon argued that although the surface o f an eccentric sphere is spherical, 104. “ Notandum etiam quod ultra motum diurnun. qui est motus primi mobilis, depre-
the sphere itself is nonuniform, as is evident from its varying thickness. Natural phi­ hensum est plures esse lationes planetarum secundum instrumenta mathematica, puta
losophers, however, insist that celestial bodies must be simple and homogeneous, and astrolabium et quadrantem, etc. Et etiam per rationem quatenus motus plmetae apparet
therefore invariant with respect to thickness. This is but another aspect o f the homo­ quandoque velocior, quandoque tardior; et planeta quandoque videtur directus, quan­
geneity argument. See Bacon, Opus tertium, 1909, 133 (for a few additional arguments, doque retrogradus, quandoque stationarius statione prima, vel secunda: quod non potest
see 132-137); Bacon, De ceiesribus, pt. 5, ch. 15, Opera, fasc. 4, 1913, 440 (and 439-443 esse secundum motum sphaerae, cum ille sit ommno umformis. Et ideo ad salvandum
for the same additional arguments). hos diversos motus Ptolemaeus et alij periti Astrologi investigaverunt circulos planetarum
102. As is evident by the assumption that the substance between orbs differs from that ot praeter sphaeras, scilicet deferentem et epicyclum, etc. Et ideo bene ait Phiiosophus quod
the orbs themselves; that the planets have different basic properties; and, as we saw in astrorum errantium plures sunt lationes quam una.” Ibid.. 448, col. !.
the preceding note, that one and the same sphere may vary in thickness.
304 THE CELESTIAL REGION
M O BILE CELESTIAL ORBS 305
One o f the few who did was Aegidius Romanus, whose discussion in
and surrounded it as being “ like marrow in a bone,” Aegidius implies that
his Hexaemeron appears to have had some influence in the seventeenth cen­
the overall heaven is hard and solid, like a bone, and that the eccentric
tury, when not only was it cited explicitly by Mastnus and Bellutus but its
cavities are filled with a soft or fluid material, akin to the marrow o f a bone.
key ideas were adopted by others.,0-i Convinced that eccentrics and epicycles
The other analogy o f blood in the veins conveys a similar relationship.
existed in the heavens and that only they could save the astronomical phe­
Aegidius asserts that the soft matter in each deferent is contiguous with
nomena, 0 10
5
1 6 Aegidius provides a brief physical interpretation for the eccentric
respect to its immediate surrounding surfaces. Thus each eccentric deferent
system. He assumes that the celestial region, from the concavity o f the lunar
is really a hollow cavity filled with a soft substance. Within each deferent
orb to the fixed stars, or eighth sphere, was one single, continuous body
orb is a spherical epicycle carried around by its deferent. The epicycle in
or orb. But just as a man or a lion is said to be one body but yet contains
turn is discontinuous, or contiguous, with respect to the eccentric. Because
within itself things that are discontinuous - such as, for example, marrow
each o f them is discontinuous, the eccentric and epicycle each has its own
in a bone, or blood in the veins - so also does the continuous single orb
proper motion by which the planet effects its retrograde or direct m otion.109
embracing the region from the M oon to the sphere o f the fixed stars contain
The relationship o f the soft matter in the eccentric deferent to the matter
discontinuities within it.107 Those discontinuities are represented by the
surrounding it is, however, left unclear. Are the two matters the same or
seven eccentric or deferent planetary orbs, which are embedded discontin-
different? The analogies indicate differences, but such an inference would
uously within the mass o f continuous matter that comprises the single orb
imply two unchangeable substances, or one unchangeable and one change­
that stretches from the Moon to the eighth sphere o f the fixed stars.
able, as Albertus Magnus had assumed earlier. Indeed, Aegidius may have
Thus did Aegidius attempt to integrate unity - the unity o f a single orb
thought o f them as the same substance, with differing densities. Once again
from the Moon to the fixed stars - and diversity - the diversity o f the seven
resorting to an analogy with animals, Aegidius observes that “just as in
eccentric deferents, one for each o f the seven planets. As he put it: “ because
animals, all parts are not equally dense, because the bone is denser than
o f the unity o f the whole body, there is one sphere and one heaven; and
flesh, so also in the heavens, all parts are not equally dense, because a star
because o f the diversity o f deferents and eccentrics, we can [also] say that
[or planet] is denser than an orb.” " 0 But whereas the rare part o f an animal
there are many spheres and many heavens.” '08 T o ensure that his readers
is more changeable than a denser part, this is untrue for the celestial region,
were in no doubt about his conception, Aegidius, as we saw, reinforced
which is unalterable and incorruptible and where every part always remains
his description with vivid analogies.
at the same level o f rarity or density.1"
Because he described the mode o f existence o f an eccentric deferent
Aegidius’s ideas about a single heaven between the Moon and fixed stars,
embedded within the continuous substance o f the single orb that contained
containing within itself eccentric deferents as hollow cavities, were adopted
105. For Aegidius’s account, see Opus Hexaemeron, pc. 2, ch. 32, 1555, 49r, col. i-54r, col. by Raphael Aversa and discussed, and perhaps adopted, by Mastrius and
1. The quotations are drawn largely from 49V, cols. 1-2 (Aegidius repeats his major Bellutus, who describe them as “ zones” (zonae) or rings (anuli)."2 They
ideas on 53v > col. 1). Duhem, Le Systeme, 1913—059, 4:110—119, gives a summary assumed a solid, single, presumably hard, starry heaven in which all the
account. For the statement about Aegidius by Mastrius and Bellutus, see note 112 of
this chapter. fixed stars and planets are embedded. Since the planets do not move them­
106. Aegidius, ibid., pt. 1, ch. 16, 1555, 15V, col. 2, says “ Advertendum etiam quod nos selves, each is carried around in its zone by an epicycle. According to Aversa,
dicimus esse in caelo eccentricos et epicyclos.” Apart from saving the astronomical
phenomena, Aegidius opposed the idea that the planets could move themselves “just as
fish are moved in water” (sicut pisces moventur in aquis). The planets are ncc self- 109. Sed habent [that is, the deferents or eccentrics) suos propnos motus per quos deferuntur
moved, because “ then there would be a division [scissio] o f the orb, or there would be planetae et epicycli, ubi sunt tixi planetae non sunt continui ipsis circulis deferentibus.
a vacuum, or two bodies would be m the same place” (quia tunc esset scissio orbis, vel Et inde est quod habent suos motus proprios per quos dicitur planeta retrogradus vel
esset vacuum, vel essent duo corpora in eodem). Ibid., pt. 2. ch. 32, 49r, col. 2. These directus.” Ibid.
same arguments also served to attack the existence o f eccentrics, as can be seen in Cecco 110. “ Advertendum etiam quod sicut in ammali omnes partes non sunt aeque densae quia os
d’Ascoli’s description o f them (see the end o f Section III and all o f Section III. 1 of this est densius carne, sic in huiusmodi coelo omnes partes non sunt aeque densae quia Stella
chapter). est densior orbe. ” Ibid.
107. “ Advertendum autem quod unum animal, ut puta unus homo vel unus leo, dicitur esse h i . “ Differunt tamen haec in animalibus et in coelo quia in animalibus partes magis rarae
unum corpus, non tamen omnia quae sunt in ipso sunt continua, ut medulla non est sunt magis passibiles, sed in coelo ita est impassibiles et ita inalterabilis et se mota ab
continua ossi, sed contigua; et sanguis non est contmuus venae, sed contiguus. Sic potest omni peregrina impressione pars rara, sicut et densa. ” Ibid.
a globo lunan usque ad octavam sphaeram; includendo ipsam sphaeram octavam did 112. Aversa, De caelo, qu. 32, sec. 7, 1627, ~2: Mastrius and Bellutus, De coelo, disp. 2, qu.
unum corpus propter continuationem totius. Non tamen omnia quae sunt in eo sunt 1, art. 2, 1727, 3:489, cols. 1-2, par. 28. Mastrius and Bellutus also mention that Aegidius
continua quia deferentes sive eccentrici non sunt continui cum huiusmodi coelo.” Ae­ seems to teach this opinion in his Hexaemeron, pt. 2, ch. 33 (Duhem, Le Systeme, 1913 —
gidius, ibid., pt. 2, ch. 32, 49V, col. 2. 1959, 4:111, correctly cites chapters 32 and 36). The term “ zones” (zonae) may derive
108. “ Propter unitatem totius corporis est una sphaera et unum coelum; propter diversitatem from John Damascene, who. as we saw earlier (note 11 o f this chapter), conceived of
autem deferentium et eccentricorum possunt dici multae sphaerae et coeli multi.” Ibid. the firmament as a single body divided into seven zones. On page 490, column 2, Mastrius
and Bellutus offer quite a different opinion (see this volume. Ch. 14, n. 134)-
M O BILE CELESTIAL ORBS 307
3o6 THE CE L E S T IA L REGION
a river traverses the whole space o f the waters up and down, once and
this single heaven is firm and solid, turning by itself from east to west with miS
again.
the daily motion as it carries all the celestial bodies with it. Within this The Coim bra Jesuits mentioned and rejected this interpretation even be­
single heaven, the seven zones, or bands, are discontinuous with the rest fore Hurtado de Mendoza w ro te ."9 If the celestial orbs are really channels
ot the heaven and are actually cavities within it.'" Although A versa denies in the sky through which the planets move by themselves, the substance
that these cavities are void, he does not indicate what fills th em ."4
that fills those channels must either be the same as the rest o f the heaven
Whereas Aegidius, Aversa, and Mastrius and Bellutus agreed that the or be o f a sublunary nature. Traditional Aristotelian arguments are invoked
planet was carried around by an epicycle, Pedro Hurtado de Mendoza, to reject both alternatives. Thus if the substance filling the interiors is o f a
according to A ve rsa,"5 assumed that the planets are self-moved rather than celestial nature, then, when the planet moves from one place to another
carried around by an epicycle. Like Aegidius and those who followed him, within the interior o f its orb, either another body succeeds it in the place
Hurtado assumed only one overall, presumably solid and- hard, heaven it iust vacated, or not. If not, then a vacuum would exist in nature, which
within which are channels that function as deferent orbs. Inside each deferent is absurd; if celestial matter succeeds into any place vacated by the planet,
orb is a planet, which is assumed to m ove by its ow n effort through its it could only do so by the processes o f rarefaction and condensation, which
orb, the interior ot which is assumed to be either void or filled with fluid. cannot occur in celestial matter. But if the interior matter is o f a sublunary
Aversa rejected the interpretation, because it has self-m oving planets and nature, the proponents o f such a theory would have assumed something
because he found a heaven that is part solid and part fluid objectionable.1,6
corruptible in the celestial region, which is also absurd.
The Coimbra Jesuits and Francisco de O viedo also rejected the concept
ot eccentric zones with self-moving planets. Citing Hurtado as a proponent
ot it, Oviedo argues'" that if the planets moved by themselves in these jo. On the assumed physical reality o f eccentrics and epicycles
celestial channels or cavities, almost the entire heaven would be hollowed
Whatever the reason for the seeming reluctance to speculate on the nature
out, because the Sun and planets do not m ove in a single track over which o f the orbs and their inner structure, it was not because o f any doubts about
they pass endlessly; rather they move over a broad band o f the sky. The
the physical reality o f eccentric and epicyclic orbs.
Sun, tor example, does not always m ove over the same path, but moves From discussions o f the three-orb system, it is obvious that those who
trom the equinoctial circle to each solstice. If Hurtado’s interpretation were accepted the “ new ” system believed in the physical reality o f material ec­
true, the Sun would require a hollow cavity that extended from one solstice
centrics and epicycles. The controversy in the Latin West was not between
to the other. The same reasoning would apply to planets like Mercury and those who argued for a system that merely saved the appearances regardless
Venus. Moreover, what tills the spaces in the cavities that are at any given o f physical reality and those who insisted that any astronomical system
moment unoccupied by the planet? O viedo thinks they would be void and must not only save the phenomena but also represent physical reality.120
that, as a consequence, we would be unable to see the stars, since the species
that enable us to see them would be untransmittable through celestial vacua. 118. “ Potest emm quodlibet astrum libere totum spatium caeli percurrere, sicuti potest piscis
Oviedo, who seems to have assumed the fluidity o f the region o f the in flumine totum aquarum spatium sursum et deorsum semel et iterum pertransire. ”
fixed stars and planets and to have treated the region as a single heaven, Ibid., 472, col. 2.
119. See Conimbricenses, De coelo, bk. 2, ch. 8. qu. 1. art. 1, 139‘L 3- 4 - In arguing tor a
denied the existence ot eccentrics and channels. Indeed he denied the ex­ fluid heaven and against solid orbs, George de Rhodes opposed the opinions ot both
istence ot any other heavens. For if the single heaven is really fluid, each Hurtado de Mendoza and Aversa. Without elaboration, he was convinced that the vast
planet must "traverse the whole space o f the heaven freely, just as a fish in13
7
*6
4 cavities which both assumed in the heavens implied the existence ot a vacuum and also
would result in collisions. De Rhodes [Dr coelo, disp. 2, qu. 1, sec. 2], 1671, 280, col.

113. Aversa mentions that some hold that Mercury and Venus do not have total orbs but 120. A tew did adopt the first alternative, including Maimonides and Thomas Aquinas.
move around the Sun by means ot epicycles (Imo plures iam sunt, qui convemunt in Sympathetic to Aristotelian cosmology but aware that it could not save certain crucial
hoc. ut Venus et Vlercurius non habeant alios proprios orbes et caelos totales. sed astronomical phenomena and also disturbed by the cosmological dilemmas inherent in
moveantur solum per epicicios circa solem). Aversa, De caelo. qu. 32, sec. 7, 1627, 72. any system o f solid eccentrics, they argued that the phenomena might perhaps be saved
114. In the third ot five opinions on the fluidity or soliditv ot the heavens, Riccioli, Almagestuin in ways that had not yet been understood or, as Maimonides put it (for Thomas, see
noinim, pars post., bk. 9- sec. i, ch. 7, 1631, 239, col. 2, describes a theorv similar to Sec. II.3 and note 35 o f this chapter), “ the deity alone fully knows the true reality, the
Aversa's. Like Aversa, he attributes it to Hurtado de Mendoza (in the latter's [De caelo. nature, the substance, the form, the motions, and the causes o f the heavens” (Maimon­
disp. 2, sec. 1 ], 1013; I have not found it there). ides, Guide, pt. 2, ch. 24, 1963, 4:327). Although what Aristotle says about the sublunar
113. Aversa, De caelo. qu. sec. 6, 1627, 63, col. 2. Oviedo also attributed this interpretation region “ is in accord with reason.” Maimonides believes that the heavens are too tar
to Hurtado (see Oviedo. De caelo. contro. i. punc. 4, 1640, 471, col. 1), as did Amicus away and coo noble for us to grasp anything “ but a small measure o f what is mathe­
(De caelo, tract. 5, qu. 3. art. 3, 1026. 284, col. 2). matical” (ibid., 326). Few in the Middle Ages shared the cosmological uncertainty
116. Aversa, ibid., sec. 7, 69, col. 1—-3, col. 2. exhibited by Aquinas and Maimonides. One who did, but went even further, was Henrv
117. Oviedo, De caelo. contro. 1, punc. 4, 1640, 1:471, cols. 1—2.
308 THE CELESTIAL REGION
M O BILE CELESTIAL ORBS 309
Rather the dispute involved a decision as to which system o f cosmic spheres
From the “ Catalog o f Questions” (Appendix I), we learn that the most
best represented physical reality - a purely concentric system or a mixture
popular question in medieval cosm ology (qu. 97) concerned the number o f
o f concentric and eccentric spheres. Repeated invocations o f dire physical
celestial spheres, “ Whether there are eight or nine, or more or less.” Scho­
consequences that might or might not follow from one or the other o f the
lastic natural philosophers faced a curious problem: should they count total
tw o rival systems serve only to confirm that medieval natural philosophers
orbs or only partial orbs? In fact, one could count either, as long as the sum
were arguing about the structure o f cosmic reality, not about convenient
was not represented as the total number o f celestial orbs. Such a move
and arbitrary arrangements o f geometric figures that might save the astro­
would have been redundant, since each total concentric orb was formed
nomical appearances. Clavius expressed traditional scholastic sentiments
from at least three partial eccentric orbs. Without the latter, there would
when he explained ([Sphere, ch. 4], 1593, 525) that eccentrics and epicycles
be no concentric orb. Pierre d’A illy makes all this quite clear when he
are not monstrous and absurd things but were adopted by astronomers for
declares that a celestial sphere is “ the aggregate o f all the orbs needed to
good reasons. Just because eccentrics have a diversity o f centers and some
save all the appearances concerning any planetary motion. In this way three
eccentrics vary in thickness should not cast doubt on them. After all, parts
eccentrics with an epicycle and with the body o f the planet are said to be
o f the Moon vary in density, as indicated by its spots. Indeed, different
only one sphere and this is how we speak about spheres in what is proposed
parts o f the heavens, not just the Moon, differ in density. W hy, then, should
here.” 123
eccentrics and epicycles be rejected because o f variations in thickness or
There were thus tw o basic ways to count orbs in the Aristotelian—Pto­
because o f a diversity oC centers? N ot until the end o f the sixteenth century,
lemaic compromise system. Duns Scotus, for example, assigned 5 eccentric
after the appearance o f the new star o f 1572 and the comet o f 1577, was
orbs to Mercury and 3 to every other planet, for a total o f 23 eccentric,
the physical existence o f eccentrics and epicycles seriously challenged.121
mobile orbs. T o this, he added 1 orb for the eighth sphere o f the fixed stars
and 1 for the ninth, or crystalline, heaven, for a total o f 25 orbs.124 O f these
IV . O n the n u m b er and order o f the m o b ile h e a v e n ly orbs 2$ orbs, at least 23 are eccentric.125 But if Scotus had counted only con­
centric, or total, orbs, he would have had only 9 orbs. Thus we may attribute
1. On the order o f the heavens to Scotus a total number o f partial, eccentric orbs, say 25, or a total number
o f concentric orbs, say 9. But w e may not speak o f their sum, or 34 cosmic
Earlier in this chapter (Sec. II.6), we emphasized the great compromise
orbs, because each concentric orb is but the sum total o f its eccentric orbs.
which produced the almost universally accepted union o f Aristotelian and
Fortunately, between the thirteenth and seventeenth centuries most scho­
Ptolemaic ideas about the relationship o f the celestial spheres. The Aris­
lastic natural philosophers left little doubt about their intentions: when
totelian—Ptolemaic fusion, which Ptolemy had already made in his Hy­
inquiring about the number o f celestial spheres in the universe, they counted
potheses of the Planets, depended on a distinction between the concept o f a
only concentric orbs, although well aware that the latter were constituted
“ total orb” (orbis totalis) and a “ partial orb” (orhis partialis), to use medieval
terminology. The total orb was a concentric orb whose center is the center
123. “ Sed tertio modo dicitur aliqua sphaera una quia est aggregatum ex omnibus orbibus
o f the earth, whereas a partial orb was an eccentric orb, that is, an orb requisitis ad salvandum omnia ilia quae apparenc circa motum alicuius planetae. Et isto
whose center is a geometric point lying outside the center o f the world modo tres eccentrici cum epiciclo et corpore planetae non dicuntur nisi una sphaera et
The concentric total orb, whose concave and convex surfaces have the ita loquendum est de sphaeris in proposito." D ’Ailly, 14 Questions, qu. 2, 1531, I49r.
This is the third ot" three ways in which d’Ailly believes the term sphaera can be used.
earth’s center as their center, is composed o f at least three partial orbs (see In the second wav, each eccentric orb o f the aggregate is counted separately, as are also
Figure 8), one o f which, the eccentric deferent, carries an epicycle in which the epicycles. The first mode includes any spherical part o f the heavens that is not
a planet is embedded. Thus were the concentric orbs o f Aristotle fused with separated from the whole o f it, a definition that also includes spherical celestial bodies,
such as stars (ibid., 148V—I49r).
the eccentric orbs o f Ptolem y.122 124. Duns Scotus, Sentences, bk. 2, dist. 14, qu. 2. Opera, 1639, 6, pt. 2:733: “ Saltern caeli
mobiles circundantes terram erunt vigindquinque, scilicet vigindtres planetarum et prae-
o f Langenstein, or Henry o f Hesse (1325-1397), who was convinced that eccentric and ter hoc caelum octavaum et caelum nonum.” In his widely used Theoricae novae plane­
epicyclic orbs were imaginary and were not to be found in the heavens. He devised a tarum, Peurbach assigned 24 orbs to the seven planets, allocating 3 to all except the
system that was “ a curious hybrid o f homocentric astronomy and an Arabic innovation Moon, which had 4, and Mercury, which had 5 (see Peurbach, Theoricae, 1987, 9—27).
introduced into Ptolemaic astronomy by Thabit ibn Qurra” (Steneck. 1976, 70). For Albertus Magnus, Metaphysics, bk. 2, tract. 2. ch. 24, Opera, 1964, 16. pt. 2:514, col.
more on Thabit’s innovation and Henry’s own system, see Steneck, 69—72. 1, describes Ptolemy’s system as one o f three proposed by “ modems.” Albertus assigned
121. Indeed, even before these two celestial phenomena appeared, Robert Bellarmine. some­ 49 orbs (he says 50, but is mistaken in his count) to Ptolemy’s system, attributing 3 to
time between 1570 and 1572, had already assumed a fluid heavens and rejected the physical the eighth sphere o f the fixed stars; 2 to the Sun; 5 to the Moon; 7 to Mercury; and 8
existence o f eccentrics and epicycles. See Chapter 14, note 75. each to Venus, Mars, Jupiter, and Saturn. Albertus's source for this strange total is a
122. On the continuity or contiguity o f these orbs, see Section III. 2 o f this chapter. mystery.
125. The eighth sphere o f the fixed stars is concentric and perhaps also the ninth sphere.
3 io THE CELESTIAL REGION MOBILE CELESTIAL ORBS 31 I

o f eccentric, or partial, orbs. The concentric orbs o f the Middle Ages and and Mercury, a number o f different planetary orders were comm only pro­
Renaissance differed from the 5$ or 49 distinguished by Aristotle in Me­ posed. Albert o f Saxony, for example, mentions four different arrangements
taphysics 12.8.1073b.2—1074a. 14. The 7 orbs assigned by Aristotle to Saturn, that had at least some support. M oving from the outermost to the innermost
for example, were all concentric (though some turned on different axes) planet, one interpretation placed the sphere o f Saturn first, then those o f
and were not counted or conceived as a single orb. By contrast, the 3 partial Jupiter, Mars, Venus, Mercury, Sun, and Moon. In another interpretation
orbs assigned to Saturn in the three-orb, Aristotelian-Ptolem aic compro­ the sphere o f Venus was placed above the Sun and the sphere o f Mercury
mise were all eccentric but could also be interpreted as forming a single below it. A third interpretation reversed the positions o f Mercury and Venus
concentric orb, because the outermost and innermost surfaces wrere con­ with respect to the Sun. A fourth interpretation placed Venus and Mercury
centric with the earth’s center. Schematic representations o f the celestial below the Sun, because it was thought more reasonable and elegant that
orbs made during the Middle Ages and Renaissance were not drawn from the Sun should be in the middle o f the planets, “ like a king in the middle
Aristotle’s cosmological system but from the concentric orbs in the three- o f his kingdom in order that the Sun should exercise its influence equally
orb compromise.12'’ above and b elow .” 128
If it was not Aristotle’s system o f concentric orbs that was incorporated O f these opinions, Ptolemy mentions the tw o which locate Venus and
into the three-orb compromise, why, then, do I designate the compromise Mercury either above or below the Sun (Almagest, bk. 9, ch. 1), thus ig­
“ Aristotelian-Ptolemaic” ? Since the compromise was ultimately the work noring the two in which Venus was placed above the Sun and Mercury
o f Ptolemy, would it not be more accurate to call it the Ptolemaic com­ belowy and vice versa. As his own choice, Ptolemy selected the order which
promise? Indeed it would. But the term Aristotelian-Ptolem aic seems more placed Venus and Mercury below the Sun, arguing that this conveniently
appropriate, because it emphasizes the most essential feature o f Aristotle’s separated three suprasolar planets, which could be any angular distance from
celestial system: the concentricity o f each planetary orb with the earth’s the Sun, from the tw o subsolar planets which could not.129 Although in the
center. It was that concentricity that enabled Aristotelian natural philoso­ Almagest Ptolemy chose not to specify whether it was Mercury or Venus
phers to embrace the system. Within the three-orb system, the order and that lay directly below the Sun, in his later Hypotheses of the Planets he locates
number o f the concentric heavens or spheres varied throughout the period Venus right below the Sun and Mercury below Venus, just above the
o f our study. All were agreed on the existence o f seven planets and the M oon .130 Thus Ptolemy favored an order o f planets in which the Sun was
fixed stars, and some included planets and fixed stars as part o f the fir­ the middle, or fourth, planet, whether counting downward from Saturn
mament.127 Here, then, was the basic core o f the celestial heavens. above or upward from the Moon below. Although this order may have
Disagreement arose, however, on the order o f the planets. Because ancient been the most popular, medieval justifications o f it rarely mentioned Pto­
and medieval astronomers could find no parallax for planets other than the lem y’s argument but rather emphasized the importance o f the Sun’s cen­
Sun and Moon, there was no way to determine the order o f the three trality, usually citing the popular metaphor o f the king in the middle o f his
superior planets. Nevertheless, Saturn was assumed farthest from the earth, kingdom .131
with Jupiter next, followed by Mars, an order that was based on the time
it took each o f them to complete its sidereal period: thirty years for Saturn, 128. “ De ordine autem talium orbium septem planetarum quidam posuerunt primo spheram
Saturni, deinde spheram Jovis, deinde spheram Martis, deinde sperarn Veneris, deinde
twelve for Jupiter, and two for Mars. Alm ost all astronomers and natural spheram Mercurii, deinde spheram solis et ultimam spheram lune. Ita quod llli posuerunt
philosophers agreed on this. The order o f Sun, Venus, and M ercury posed Venerem et Mercurium supra solem. Alii autem posuerunt Venerem supra solem et
quite different pioblems. Without detectable parallaxes, and because Venus Mercurium inFra; alii autem econverso. Sed quicquid de hoc sic rationabilius esse videtur
quod tres planete sint supra solem et tres inFra et sol in medio tanquam rex in regm
and Mercury always remained in the vicinity o f the Sun. the order o f these medio ad Finem quod supra et inFra possit equaliter influere et illuminare.” Albert ot
three planets was not determinable except on the basis o f arbitrary, non- Saxony, De celo, bk. 2, qu. 6, 1518, 105v. col. 1. Melanchthon (1550, 51V) described
astronomical reasons. The Moon, however, was universally assumed to be this order as “ the oldest and common opinion” (Sequamur lgitur vetustissimam et
communem sententiam, quae medium locum Soli tribuit, sitque hie ordo: Saturnus,
the closest planet to earth. Iupiter, Mars, Sole, Venus, Mercurius, Luna). For the widespread use oF "the king in
Because o f the various combinations possible between the Sun, Venus, the middle o f his kingdom” as a metaphor For the Sun, see Chapter 11, note 28.
129. In book 9, chapter 1, Ptolemy speaks oF the Five planets, thus excluding the Sun and
Moon. Perhaps he chose to exclude the Moon From the subsolar company oF Venus and
!2(>. For an 1 i-orb version, see the diagram in this chapter From Peter Apian’s Cosmographiais Mercury because the Moon could be any angular distance From the Sun. The Moon, ot
liber (Figure 9) and also see Reisch, Margarita philosophies 1517. 244; an 8-orb version course, had to be, and was aUvavs, counted among the planets below the Sun.
appears in Paris, Bibliotheque Nationale, fonds latin, MS. 6280, 2or (12th century), 130. Van Helden, 1985, 20-23, has a detailed discussion.
which is reproduced in Grant, 1978a, 276. 13 1. See Chapter 11, Section 1.3 and note 28. Although Copernicus emphasized the Sun’s
127. One who did not was Clavius, who equated the Firmament solely with the sphere ot centrality in quite another way, by having all the planets revolve around it, the Sun’s
the Fixed stars. See Clavius, Sphere, ch. 1, Opera, 1611, 3:23. importance as the middle planet was heavily emphasized in scholastic cosmology.
M O BILE CELESTIAL ORBS 313
31 2 THE CELESTIAL REGIO N
closest when they are below the Sun, it follows that no fixed order exists,
Despite certain criticisms o f Ptolem y’s order o f the planets,'32 most fol­
but that Mercury, Venus, and the Sun vary their order with respect to the
lowed it,13 133 although they could have chosen another opinion (the first
2
earth. Although the concentric orbits that Martianus clearly attributed to
mentioned in my discussion o f Albert o f Saxony), one that placed the Sun
Mercury and Venus would be altered by the ninth century - indeed, they
below Venus and Mercury (that is, V enus-M ercury-Sun; or, with the po­
would usually be depicted as intersecting - the varied order o f Mercury,
sitions o f Venus and Mercury reversed, M ercury-Venus—Sun), leaving onlv
Venus, and the Sun remained.136*B y the seventeenth century, this variable
the Moon below it. Clavius called this the “ Egyptian system ,” as it was
arrangment o f the three planets was known as the “ Capellan” system and
known traditionally, and cited both Plato and Aristotle as supporters.134
was taken as the basis o f Tycho Brahe’s own Sun-centered system involving
A trace o f another kind o f planetary “ order” should be mentioned, one
five planets. The Capellan system had numerous supporters, many o f whom
that probably derives from The Marriage oj Mercury and Philology o f Mar-
were Jesuits (Schofield, 1981, 172-183).
tianus Capella, who may have composed it sometime between 410 and 439.
The varied order o f the inferior planets would appear again in the four­
Martianus was probably the first extant Latin author to have adopted an
teenth century in Jean Buridan’s Questions on De caelo. In Buridan’s discus­
arrangement o f the planets wherein Mercury and Venus are assumed to
sion, however, the orbits o f Venus and Mercury are neither concentric nor
orbit around the Sun, rather than the earth, for which reason Copernicus
intersecting, but eccentric and epicyclic. Observing that the inferior planets
mentioned Martianus as a precursor.'33 B y proposing Sun-centered orbits
traverse their orbits in less time than the superior planets, Buridan asks why
for Mercury and Venus, Martianus could not have accepted a single, fixed
Mercury, Venus, and the Sun seem to complete their orbits in the same
order o f the planets but was committed to a variable order, such that when
time, something no other planets do. In response to his own question,
Mercury and Venus are above the Sun with respect to the earth (that is,
Buridan declares:
farthest from the earth) their descending order is Venus-M ercury-Sun, but
when they are below the Sun (or nearest the earth) their descending order
some reply that this occurs because these three planets are fixed in the same sphere,
is Sun-M ercury—Venus (Eastwood, 1982, 147). Because M ercury is closest
although they have different epicycles and eccentrics within it. And this is probable
to the earth when the two inferior planets are above the Sun and Venus is
because, as the Commentator [i.e., Averroes] says, many ancients assumed that
Venus and M ercury were above the Sun, [while] others [assumed that] they were
132. Albert ot Saxony, De celo, bk. 2, qu. 6, 1518, 105V, col. 1, reports a counterargument below the Sun. [N ow] this could be [true], because when they are in the a u x e s o f
based on lunar eclipses. If the Moon, which lies between us and the Sun, can eclipse the
Sun, why do not Venus and Mercury also eclipse it? Some explain this by the greater their eccentrics and epicycles, they are higher than [or above] the Sun; and [when
transparency o f Venus and Mercury, which permits the Sun’s rays to penetrate them. they are] in the opposite o f the a u x e s , they are lower than [or below] the Sun.” 7
The lesser transparency o f the Moon blocks the Sun’s rays and creates an eclipse. A
second explanation is based on the principle that the closer an opaque body is to us, the
greater the eclipse it can cause as compared to a body, or bodies, that are more remote. Buridan’s description o f the variable order o f Mercury and Venus with
But Venus and Mercury are so far away from us that the part o f the Sun that they eclipse respect to the Sun seems equivalent to assigning them Sun-centered orbits
is not visible to us. In his Hypotheses of the Planets, Ptolemy suggests that the smallness within the frame o f the Ptolemaic system o f eccentrics and epicycles. To
o f Venus and Mercury may explain why so few o f their transits across the Sun have
been observed (see Van Heiden, 1985, 21). achieve these orbits, someone had boldly proposed that Mercury, Venus,
133. For example, Clavius, Sphere, ch. 1, Opera, 1611, 3:43. Additional supporters are men­ and the Sun be encompassed within a single sphere, although each planet
tioned in Chapter 11, Section 1.3 and note 28 o f this volume. would have its own eccentric deferent and epicycle. Buridan characterized
13-1- For Plato, Clavius mentions the Timaeus, and for Aristotle he cites De caelo, book 2,
chapter 12, and Meteorology, book 1, chapter 4, in neither o f which does Aristotle present this arrangement as “ probable” (probabile) because it reconciled the differ­
an order o f planets (see Clavius, Sphere, ch. r. Opera, 1611, 3:42). Indeed, Aristotle does ences o f opinion mentioned by Averroes, namely that some ancients placed
not even mention an order o f the planets in his famous discussion on the number ot Mercury and Venus above the Sun while others located them below the
spheres in Metaphysics 12.8.1073b. 18—1074a. 14. However, Clavius also mentions Aris­
totle’s De mundo, suggesting it might have been talsely attributed to Aristotle, as indeed Sun.'38 The interpretation Buridan describes saves both alternatives.
it was. In the pseudo-Aristotelian De mundo 3923.20—30 (see Aristotle [Forster], 1984)- If his report for a variable order for the Sun, Mercury, and Venus was
the author does indeed present the Egyptian order o f the planets, with Mercury, Venus,
and Sun in descending order. 136. Eastwood, 1982, 149—155, describes the transformation o f Martianus’s text.
[35. After rejecting Vitruvius. Chalcidius, and Macrobius as Latin sources for Sun-centered 137. “ Aliqui respondent quod hoc est quia illi tres planetae fixi sunt in eadem sphaera, licet
planetary motion, Eastwood (1982, 146, n. 1) argues that Martianus was the first Latin in ea habeant diversos epiciclos et diversos eccentricos. Et hoc est probabile quia sicut
author to propose heliocentric orbits for Mercury and Venus, a theory that Martianus dicit Commentator, multi antiqui posuerunt Venerem et Mercurium supra solem, alii
may have derived ultimately from Theon o f Smyrna (fl. ca. 130; Eastwood, ibid., G 1!- autem infra solem; quod poterat esse quia quando erant in augibus eccentricorum et
For Copernicus’s citation o f Martianus, see Copernicus, Revolutions, bk. 1, ch. i° epiciclorum suorum, tunc erant altius quam sol, et in opposito augium erant bassius.”
[Rosen], 1978, 20. Eastwood also shows (1992, 233, 256) that no evidence exists m Buridan, De caelo, bk. 2, qu. 20, 1942, 220.
support o f the claim that Heraclides o f Pontus (4th c. a . d .) had previously proposed 138. Buridan does not say where Averroes made this remark.
heliocentric orbits for Mercury and Venus.
314 THE CELESTIAL REGION
MOBILE CELESTIAL ORBS 315
derived ultimately from Martianus Capella, Buridan gives no indication of
it. By contrast, Capella’s name is the only one mentioned by Copernicus, 2. The number o f orbs
although he also alludes to “ certain other Latin writers.” ' 39 Is it possible
(though perhaps unlikely) that Copernicus had in mind one or more Latin a. Do orbs exist beyond the eight orbs of the planets and fixed stars?
writers o f the late Middle Ages, perhaps Buridan himself? And who did
Whatever the order chosen, all were agreed that the planets and fixed stars
Buridan have in mind when he speaks o f “ some” (aliqui) who placed the
together accounted for at least eight concentric orbs or heavens. Cosm ol-
Sun at the center o f the orbits o f Mercury and Venus and encompassed the
ogists and astronomers were, however, soon compelled to decide whether
three planets within a single eccentric-epicyclic sphere? Is it plausible to
any spheres existed beyond the fixed stars and if so, whether they were
suppose that Buridan and Copernicus might have included one or more of
mobile or immobile. With respect to mobile orbs, responses usually de­
the same individuals? These are questions to which answers seem unlikely.
pended on the number o f motions assigned to the sphere o f the fixed stars.
Indeed, we cannot even propose a definitive reply to the question of
Throughout the Middle Ages, at least three motions were attributed to it:
whether Buridan was himself a supporter o f Sun-centered orbits for Mer-
(1) a daily motion from east to west; (2) a precession o f the equinoxes o f
curv and Venus. For although he calls that interpretation “ probable” (prob-
1 degree in 100 years, producing a complete revolution o f the starry sphere
abile), Buridan introduces an alternative opinion by observing that Ptolemy
in 36,000 years; and (3) a progressive and regressive motion o f the stars
adopted the fixed order o f Sun-Venus-M ercury, where the Sun is farthest
known as “ access and recess,” or “ trepidation,” a theory proposed by the
from the earth and Mercury nearest, and each planet has its ow n sphere.
ninth-century Arab astronomer Thabit ibn Q urra.144 Although Thabit’s
To explain how these three planets complete their independent revolutions
trepidation theory was intended as a substitute theory for the precession o f
in the same time, we would have to assume, says Buridan, “ a similar ratio
the equinoxes, not as an additional motion, many natural philosophers
o f moving intelligences to moved spheres.” '40 Which o f the two alternatives
treated them as two distinct motions, as we shall see.
Buridan favored is thus left unclear.'4'
If we accept the principle that every motion requires its own separate
What emerges from Buridan’s discussion that is o f considerable signifi­
sphere, a sphere would have to be added for every such motion. But this
cance is the fact that he reported sympathetically a limited heliocentric
did not signify that a sphere could not move with multiple motions. To
system involving Mercury and Venus, encompassed within a system ot
understand this, we must realize that medieval natural philosophers distin­
eccentrics and epicycles. Since Copernicus used Ptolemaic eccentrics and
guished between the “ proper” motion o f a sphere and those motions that
epicycles, Buridan seems thus far to have left us the first extant, unequivocal
were imposed upon it externally from the motions o f superior spheres. A
description o f limited heliocentric orbits within a system o f eccentrics and
sphere could have only one proper motion, usually characterized as a simple
epicycles.144 Echoes o f it were still heard in the seventeenth century.13143
2
0
14
9
motion. Thus the eighth sphere o f the fixed stars might have the motion
o f precession as its proper motion. But it also had a daily motion and a
139. Copernicus, Revolutions, bk. 1, ch. 10 [Rosen], 1978, 20. Among “ other Latin writers,” motion o f trepidation. The sources o f these two motions had to be sought
Rosen suggests that Copernicus may have included Vitruvius, Architecture, IX, 6 (Cop­
ernicus, ibid.. 358). Although Eastwood has eliminated Vitruvius, Macrobius, and Chal- in orbs that were independent of, and distinct from, the eighth sphere. On
cidius as real believers in Sun-centered planetary motion, it does not follow that the universal principle that “ no sphere is ever moved with the motion o f
Copernicus and others would have viewed earlier, potentially relevant Latin authors in an interior sphere but is moved with the motion o f a superior sphere,” as
the same light. In his Almagestum novum, Riccioli considered the Egyptians as the in­
ventors o f the Capellan system and, in addition to Capella, also named Vitruvius, Pierre d’A illy put it,'45 the other two motions o f the eighth sphere were
Macrobius, and Bede as its supporters (see Schofield, 1981, 173 and 347, n. 23).
140. Here are Buridan's words on Ptolemy: “ Dicitur tamen quod Ptolomeus geometrice
invemt sphaeram solis esse supra sphaeram Veneris et sphaeram Veneris supra sphaeram noreworthv, however, that he makes no mention o f the Sun-centered orbits o f Mercurv
Mercurii. Et tunc causa propter quam illae sphaerae sic aequali tempore perticerent suas and Venus.
circulationes, esset similis proportio intelligenciarum moventium ad sphaeras motas. 143. For Amicus’s discussion, see Chapter 14, Section VIII.2.c.
Buridan, De caelo, bk. 2, qu. 20, 1942, 220, lines 28-33. On the relationships ot intel­ 144. The theory o f trepidation arose from discrepancies in the observation o f precession (see
ligences to the spheres they move, see Chapter 18. Section II. Campanus o f Novara, Theorica planetarum. 19^ 1, 378—379. and Drever. 19^3, 276—2~7).
141. Although in his Questions on the Metaphysics (bk. 12, qu. 10, 1518, 73r~74r), Buridan Ordinarily, either precession or trepidation should have been employed, but some scho­
concludes that all astronomical appearances can be saved by eccentrics alone without lastics assigned both motions to the stars. Albertus Magnus, Metaphysics, bk. 2, tract.
epicycles (tor my translation, see Grant, 1974, 524—526), he speaks ot both eccentrics 2. ch. 24, Opera. 1964, 16. pt. 2:514, col. 1, attributes all three motions to the eighth
and epicycles in the heliocentric argument about Venus and Mercury. Nevertheless, he sphere and mentions Thabit, or Thebit. as the discoverer o f trepidation, which he calls
calls the latter argument “ probable” and raises no objections to the inclusion ot epicycles. “ the motion ot accession and recession” (motus accessions et recessionis). or "pro­
I am ignorant o f the order o f composition o f Buridan’s Metaphysics and his De caelo. gression and regression."
142. Albert o f Saxony considered the same question (De celo, bk. 2, qu. 16, 1518. m v , col. 145. This is the third ot tour assumptions, where d’Ailly says: "Tertio suppomtur quod aliqua
2—U2r, col. 2) as did Buridan and included most o f what Buridan discussed. It is sphaera nunquam movetur ad motum sphaerae inferiors sed bene ad motum sphaerae
superiors.” D ’Ailly, 14 Questions, qu. 2. 1531, 149s
MOBILE CELESTIAL ORBS 317
316 THE CELESTIAL REGION

quite naturally attributed to superior spheres: one motion to a ninth sphere •fcrimapara. £o!.<5.
and the other to a tenth sphere. Thus with three motions assigned to the
eighth sphere, it was usual to add two spheres; if four motions were assigned
to the sphere o f the fixed stars, three additional orbs were customarily added.
The spheres that allegedly existed beyond the eighth sphere were assumed
to be devoid o f celestial bodies and therefore w holly transparent and in­
visible.
For those who assumed only an east-to-west daily motion for the stars,
no additional orb was necessary; eight movable orbs sufficed.146 Others,
for example, Peter o f Abano, argued for nine spheres. Peter assumed two
motions for the sphere o f the fixed stars: one, the daily motion, he assigned
to the eighth sphere; the other, the precession o f the equinoxes, he attributed
to the ninth sphere.'47 In a somewhat different arrangement, Illuminatus
Oddus ([De coelo, disp. 1, dub. 14], 1672, 41, col. 2) assumed nine mobile
heavens, equating the ninth with the primum mobile and crystalline orb. But
Albert o f Saxony, Roger Bacon, Them on Judaeus, and Pierre d’Ailly,
among others, attributed all three motions to the fixed stars and therefore
assumed the existence o f ten mobile orbs.14814 0Thus Albert assigned the daily
5
9
motion to a tenth orb, the primum mobile; a motion o f trepidation to the
ninth sphere; and the motion o f precession to the eighth or starry sphere.'49
Clavius characterized the theory o f ten mobile orbs as “ the most cele­
brated that has appeared in the schools o f astronomy to this day” ' 50 but did
not himself adopt it, while the Coim bra Jesuits defended it as their opinion,
observing that not only astronomers, but also many Peripatetic philosophers
had embraced it.'5' Both Clavius and the Conimbricenses mention that a

146. Albert o f Saxony, De celo, bk. 2, qu. 6, 1518, 105V, col. 1, explains: “ Ulterius sciendum
est quia aliqui philosophi non perceperunt octavam spheram moved pluribus modbus figu re 9. The movable celestial spheres, ranged in order from the lunar
sed unico, scilicet ab oriente in occidentem, dixerunt spheram octavam esse ultimam et orb to the “ first movable heaven” (prim um m obile). Encompassing the
nullam esse ultra.” D ’Ailly, 14 Questions, qu. 2, 1531, I49r, mentions the opinion ot whole is the immobile empyrean heaven, “ dwelling place o f God and
those who insist on the existence o f only eight orbs. They say that the eighth orb moves all the elect.” (Peter Apian, Costnographicus liber (1524), col. 6. Courtesy
with only one motion (the daily motion) and that there is no need to assume a ninth
Lilly Library, Indiana University, Bloomington.)
orb. D ’Ailly rejects this opinion and observes that astronomers (astrologi) deny the at­
tribution o f only a single motion to the eighth sphere. Amicus, De caelo, tract, a, qu.
6. dubit. 2, 1626, 185, col. 2, also mentions the eight-orb interpretation and observes ten-orb scheme was accepted in a similar form by such great astronomers
that its proponents reject the motion o f trepidation.
147. See Peter o f Abano, Lucidator, differ. 3, 1988, p. 217, lines 21-22 through p. 218, line
as Albategnius (al-Battani),'5~ Thabit ibn Qurra, King Alfonso o f Spain,
2. Peter specifically argues against those who assume ten spheres. Georg Peurbach, and Regiomontanus, all o f w hom applied these three
148. Albert o f Saxony, De celo, bk. 2, qu. 6, 1518, io6r [incorrectly foliated as 107], col. u motions to the eighth, ninth, and tenth spheres.'53 Thus they assigned the
d’Ailly, 14 Questions, qu. 2, 1531, I49r; and Bacon, De celestibus, pt. 4, ch. 3, Opera,
fasc. 4, 1913, 388; ibid., pt. 5, ch. 18, 447, 449; ibid., ch. 19, 455; for Themon Judaeus,
daily motion from east to west to the tenth sphere, which carried with it
see Hugonnard-Roche, 1973, 105. the eighth sphere o f the fixed stars and all the inferior planetary orbs; to
149. Although d’ Ailly assigned motion o f precession as the proper motion o f the eighth the ninth heaven they attributed a west-to-east motion that carried with it
sphere, he does not specify the motions o f the ninth and tenth orbs. It was usual, however,
to assign the dailv motion to the outermost moving orb, or primum mobile. It is therefore non solum Astronomi, quorum observatio et experientia hac in re fidem meretur, sed
likely that d’Ailly followed Albert o f Saxony and assigned the daily motion to the tenth etiam multi e Peripatetica schola nobiles Philosophi amplectuntur. ” Conimbricenses, De
orb and trepidation to the ninth orb. Clavius reports a different arrangement (see the coelo. bk. 2, ch. 5, qu. i, art. 1, 1598. 247.
next paragraph). 152. Only the Coimbra Jesuits mention Albategnius.
150. ‘‘Hie igitur denarius numerus orbium coelestium in scholis astronomorum celebernmus 153. The same interpretation is also reported, though not accepted, by Amicus, De caelo,
ad hanc usque diem extitit.” Clavius, Sphere, ch. 1, Opera, 1611, 3:23. tract. 4, qu. 6, dubit. 2, 1626, 185, col. 2—186, col. 1.
15 1. “ Haec igitur sententia de denario coelestium sphaerarum numero nobis probatur quaih
318 THE C EL EST IA L REGION
M O BILE CELESTIAL ORBS 319
the firmament, or sphere o f fixed stars, and all the inferior planetary orbs;
carried no star or planet. With this in mind, Clavius posed an interesting
and finally, to the eighth orb o f the hxed stars they assigned the motion o f
objection against himself. He explains that in the twelfth book o f his Meta­
trepidation.'54 The functions assigned to the eighth to tenth orbs could vary,
physics, Aristotle insisted that every motion o f an orb directly represents
as when some assigned the motion o f precession to the ninth orb rather
the motion o f a star or planet, or, as the Conimbricenses explain it ([De
than to the eighth o rb .'55
coelo, bk. 2, ch. 5, qu. 1, art. 3], 1598, 250), Aristotle argued that a celestial
Clavius moved from the ten-orb system to one with eleven mobile orbs.
motion exists for the sake o f a planet or star. Since no star or planet exists
In this, he was influenced by Copernicus, whom he called “ a most learned
in the ninth through eleventh orbs, would this not signify a vain and su­
man and most praiseworthy” (vir longe doctissimus, omnique laude dig-
perfluous existence? Clavius replies that although no star or planet exists in
nissimus). According to Clavius, Copernicus assigned four, not three, mo­
those orbs, the motion o f each such orb, as we have seen, exercises a direct
tions to the fixed stars.15 1156 In apparently following Copernicus, Clavius
4
influence on a planet or star that exists in another heaven. The Coimbra
assumed three mobile orbs beyond the eighth sphere o f the fixed stars and
Jesuits argued further (ibid., 251) that orbs do not exist solely to cause the
assigned the four motions to orbs eight to eleven. T o the eleventh orb, he
motion o f planets or stars, as is evidenced by the common assumption that
assigned the daily motion from east to west; to the tenth orb, a motion
all parts o f an orb - even those parts that are distant from the celestial body
from north to south and south to north; to the ninth orb, he assigned “ a
it carries - exercise an influence on inferior bodies below the Moon.
certain unequal libration from east to west and west to east” ; and to the
Clavius was apparently satisfied with his response to the objection but.
eighth sphere o f the fixed stars, a proper motion from west to east, which
like all o f his predecessors and contemporaries, ignored the fact that the
appears to represent the precession o f the equinoxes.157
extra orbs simultaneously affected a given celestial body or bodies. Thus
Although Aristotle had assigned multiple orbs to each planet (but only
the three or four orbs (i.e., orbs nine to eleven or twelve) assigned to
one to the fixed stars), he devised a system in which the orbs o f one planet
represent each o f three or four motions assigned to the eighth sphere o f the
could not effect the orbs o f another. M oreover, since the planet itself was
fixed stars act continuously and simultaneously on the eighth orb. The fixed
carried by only one o f the multiple orbs assigned to it, the remaining orbs
stars would thus be subject to three or four simultaneous motions, which
is what was supposed to be avoided by invoking one orb for each motion
154. Since both texts are substantially similar, I shall cite only that o f Clavius: “ Post Ptole- o f any celestial body. If one were merely saving the phenomena geometri­
maeum deinde. anms mteriectis M CX L fere, Tebith, Alphonsus Hispanorum rex Anno
Domini M CC L, Georgius deinde Peurbachius et Ioannes de Regiomonte insignes as-
cally, the number o f spheres assigned to represent the total number o f
tronomi, deprehenderunt quidem in stellis fixis duos motus praedictos, sed eas praeterea motions would be irrelevant and without physical consequences. But for
observarunt tertio quodam motu, quern accessus et recessus dixerunt, ut paulo post Clavius and other astronomers and natural philosophers, the orbs were
declarabitur agitari. Quare cum corpus simplex unico tantum motu fern sit aptum, ut
volunt Philosophi non potest nonum coelum esse pnmum mobile, sed supra ipsum erit assumed physically real, and all motions were real motions. In the end, the
aliud statuendum coelum quod sit primum mobile. Ita enim fiet ut decimum hoc coelum fixed stars would be subject to contrary motions, since the orbs above would
motu diurno quern habet proprium ab oriente in occidentem, secum trahat omnes coelos act simultaneously on the orb o f the fixed stars. Such problems were ap­
inferiores atque adeo Firmamentum quoque cum stellis fixis spacio 24 horarum; nonum
deinde coelum circumvehat suo proprio motu quern obtinuit ab occidente in onentem parently ignored. Indeed, once the multiple motions o f a planet or the stars
et Firmamentum et reliquos omnes coelos infra ipsum; octavum denique coelum, seu were assigned to independent orbs, the problem was considered resolved.
Firmamentum, in quo stellae fixae existunt, moveatur tanquam proprio motu, accessu At least one medieval natural philosopher opposed the assumption o f a
illo et recessu, quern praefati astronomi repererunt.” Clavius, Sphere, ch. 1, Opera. 1611,
3:23. The Conimbricenses declare that the motion o f access and recess is also called ninth sphere beyond the eighth. Nicole Oresme, noting that astronomers
trepidation (“ Et denique firmamentum motu sibi proprio moveatur, accessu illo et re­ had determined that the eighth orb o f the fixed stars had a movement
cessu, quern titubationis, seu trepidationis, motum vocant"). Conimbricenses, De coelo, composed o f several different motions, for which reason they assumed the
bk. 2, ch. 5, qu. 1. art. 1, 1598. 247.
155. As reported by Amicus, De caelo, tract. 4, qu. 6, dubit. 2, 1626, 185, col. 2.
existence o f a ninth sphere, thought such a move superfluous. Unfortu­
156. “ Nostra denique tempestate Nicolaus Copernicus, vir longe doctissimus, omnique laude nately, Oresme rests content merely to inform his readers that in an earlier
dignissimus, non solum tres in stellis fixis motus observavit, sed quatuor." Clavius, w ork he had explained how two different, simultaneous motions could be
Sphere, ch. 1, Opera. 16r 1. 3:23.
157. “ Nam ad motum undecinn coeli, seu primi mobilis, moventur omnia astra ab ortu in assigned to the starry orb without invoking a starless ninth orb .158
occasum; et ad motum decimi coeli a septentrione in austrum et ab austro in septentri- But if we are ignorant o f Oresm e’s way o f avoiding a ninth orb, there
onem per 24 minuta sub coluro solstitiorum; ad motum vero noni coeli habent libra- were apparently other suggestions for obviating the need for additional orbs
tionem quandam inaequalem ab ortu in occasum et ab occasu in ortum sub ecliptica
decimae sphaerae per minuta 140; motu denique proprio octavi orbis stellae fixae cir- beyond the eighth. Albert o f Saxony reports one in which two intelligences
cumvehuntur ab occasu in ortum.” Ibid., 23—24. Amicus, De caelo. tract. 4, qu. 6, dubit. are assigned to the eighth orb, one to move it from east to west, the other
2, art. 4, 1626, i88, col. 2, rejects the eleven-orb system, believing that an eight-orb
system is more probable.
158. Oresme, Le litre dti del, 1968, 488-491. Oresme does not name the earlier treatise.
M O BIL E CELESTIAL ORBS 321
320 THE CELESTIAL REGION

to move it from west to east, a solution that Albert rejects because it violates Although some Greeks thought o f the celestial region as composed o f
Aristotle’s dictum that to one and the same orb only one motive intelligence the four elements, which therefore included water, most would have denied
can be assigned.'59 One might also argue for the existence o f only eight the existence o f a large mass o f water beyond the fixed stars. Indeed, Ar­
orbs by assuming that the eighth orb does not itself move with a plurality istotle denied the possible existence o f water beyond the concave surface o f
o f motions - three, to be precise. Thus tw o o f the motions most frequently the lunar sphere.'63 Biblical exegesis, however, demanded that the waters
assigned to the eighth orb o f the fixed stars might be assigned to the earth, above the firmament be conceived as real, although their precise nature was
namely a west-to-east motion, which would account for the daily motion open to debate.
o f the heavens from east to west, and the progressive and regressive motion Relatively early in the history o f Christianity, those waters were con­
o f trepidation, which Thabit ibn Qurra had discovered.15 160 In this scheme,
9 ceived as crystalline, a term which, as we saw earlier, was sometimes
only the motion o f precession is assigned to the eighth sphere.'6' Because thought o f as applying to fluid waters and sometimes to waters that were
there is no direct evidence to demonstrate that the earth moves in these congealed and hard like a crystal. The latter gained support from Ezekiel
ways, Albert expresses reservations about this interesting opinion, although 1.22, which speaks o f an awesome crystal stretched like a vault over the
he then tantalizingly suggests, without elaboration, that one might devise heads o f the animals o f the firmament.164*Thus for Saint Jerome and Bede
a way to avoid the difficulties.162 the waters above the firmament were conceived as crystal-like, which sig­
nified hardness, whereas for Saints Basil, Gregory o f Nyssa, and Ambrose
they were fluid. Whether fluid or hard, however, during the early Middle
b. The theological heavens: the firmament and the crystalline orb Ages, say from the fifth to the mid-twelfth century, the crystalline orb was
The sacred text o f Genesis 1.6 demanded that waters o f some kind be usually located above the sidereal heaven, or firmament o f fixed stars, and
assumed to lie above and beyond the heaven o f the firmament, or fixed below the empyrean heaven.'65
stars, thus giving rise to tw o theological heavens or orbs: the firmament Whether tw o or three in number, the starless and planetless orbs were
and the crystalline orb. Earlier in this study (Ch. 5, Secs. VI-VII), we assumed to be not only material, physical entities but also transparent and
considered some o f the essential features o f these two orbs. We saw that invisible. It was therefore easy to identify one or all o f them with the waters
the firmament (Jirmamentum) was given a number o f interpretations. A few above the firmament, or the crystalline sphere, as those waters were often
associated it with the air beneath the heavens; others identified it with the described. Some considered the identification o f waters with one or more
region between, and including, the M oon and the fixed stars; while some orbs above the firmament, or above the eighth sphere o f the fixed stars,
limited it solely to the eighth orb o f the fixed stars. Either o f the last two essential on theological grounds,'66 or at least viewed it as a sphere that was
interpretations was compatible with the three-orb compromise system. For named “ crystalline” by theologians.167 Because o f their “ clarity and trans­
even if one counted all the planets and stars as comprising the firmament, parency,” Clavius identified the ninth through eleventh orbs with the crys­
the planets were still conceived as subdivided into seven orbs, while the talline sphere.168 For the same reason, as well as for the freezing power that
fixed stars were embedded w holly in an eighth sphere; or one could simply they allegedly have, the Coimbra Jesuits identified the ninth and tenth
assign the term “ firmament” solely to the eighth sphere o f the fixed stars. mobile orbs with the watery, or glacial, heaven, above the firmament, which
O n either interpretation, the end result was identical. The waters above the is usually characterized by one common name: “ crystalline.” '69 Occasionally
firmament, however, required a somewhat more complex interpretation
163. See Duhem, Le Systeme, 1913-1959. 2:488.
before they could be assimilated into the secular cosm ology o f the Aris­ 164. “ Et similitudo super capita animalium firmamenti, quasi aspectus crystallis horribilis,
totelian—Ptolemaic compromise system. et extenti super capita eorum desuper.” See Bible (Vulgate), 1965; also Campanus of
Novara, Theorica planetarum, 1971, 393-394, n. 54.
159. Albert o f Saxony, De celo, bk. 2, qu. 6, 1518, 105V, col. 2—106 [mistakenly foliated 165. Thomas Aquinas [Sentences, bk. 2, dist. 14, qu. 1. art. 4], 1929-1947, 2:354, cites Rabanus
Maurus (ca. 776-856) as one who assumed seven heavens: empyrean (empyreum), crys­
I07]r, col. 1.
160. “ Aliter potest sustineri quod non essent nisi octo orbes et quod octava sphera non talline (chrystallimim), sidereal (sidereum), fiery (igneunt), olympian (olympium), ethereal
moveretur pluribus modbus sed quod ipsa apparet moveri pluribus motibus est ex eo (aethereum), and airy (aereum).
quod terra movetur ab occidente in orientem et per unum alium motum terre possit 166. See Campanus of Novara, Theorica planetarum, 1971, 183, and Clavius, Sphere, ch. 1,
salvare apparenda motus accessus et recessus octave sphere que invenit Thebit.” Ibid.. Opera, 1611, 3:24.
167. Michael Scot declares (Sphere, 1949, 283): “ Secundum celum dicitur nona sphera que a
106 [mistakenly foliated I07]r, col. 1.
161. For further discussion o f the earth’s possible axial rotation, see Chapter 20, Section V. theologis dicitur cristallinum.”
162. “ Sed istud non videtur esse omnino tutum quia non apparet prima facie quid terrain sic 168. Clavius, Sphere, ch. 1, Opera, 1611, 3:24.
moveret. Nihilominus forte qui niteretur in defensionem huius opinionis posset excog- 169. “ Nonum et decimum, quos, ut alibi retulimus, theologorum nonnulli significari putant
itare faciliter modum hoc evadendi et plura alia dictam opimonem multum colorantia. in sacris literis nomine aquarum cum Geneseos 1., dicunt Deum aquas ab aquis interposito
Albert o f Saxony, De celo, bk. 2, qu. 6, 1518, 107 [really io6]r, col. 1. firmamento secrevisse. et in Psalmo 148, cum dicuntur aquae esse super coelos aquae,
322 THE CELESTIAL REGION
MOBILE CELESTIAL ORBS 323
there was hesitation in such identifications. Convinced on theological
it was not the heaven created on the first day with which Christian natural
grounds that there was a crystalline sphere, Campanus o f Novara was
philosophers and theologians eventually identified the ninth orb - and even
initially uncertain about identifying it with the ninth orb, though he even­
tenth and eleventh orbs. It was rather with the waters above the firmament,
tually did so.170 Peter o f Abano went further and insisted that those who
which they often described as a crystalline orb.
theologized more spheres by “ assuming a crystalline, or aqueous [sphere]
In these brief and general descriptions and allusions to a ninth sphere, we
and an empyrean, or fiery [sphere]” did so on the basis o f revelation, not
learn o f certain significant features. During the thirteenth century, some
reason or experience, on which he based his w ork.' 1
authors - Robertus Anglicus and Michael Scot, for example - upheld the
The concept o f a ninth orb probably entered western Europe within the
existence o f a ninth orb, despite a degree o f uncertainty. Robertus admits
corpus o f Greco-Arabic astronomy and cosm ology that was translated from
that he was aware o f no authority who had demonstrated the necessity for
Arabic into Latin. Thus Roger Bacon held that in some translations of
more than eight orbs. Nevertheless, because two motions were associated
Aristotle’s Metaphysics a ninth orb is assumed,172 as it was in works by
with each o f the eight planetary orbs, as well as with the eighth starrv orb,
Thebit and Al-BitrujI (Alpetragius).171 The introduction o f a ninth orb was
he thought it reasonable to assume that a ninth orb existed which possessed
made within a secular and astronomical context, having no connection with
only a single, simple motion - presumably the daily east-to-west motion.
theology. A ninth orb was therefore sometimes, and perhaps even often,
But the ninth orb exercised a cosmic influence not only by motion but also
mentioned without any reference to the waters above the firmament or the
by light, which was evenly distributed over the orb, in contrast to planet­
crystalline sphere. It was the first astronomical sphere beyond the sphere
bearing orbs where light was concentrated around and near the celestial
o f the fixed stars and was usually assigned one o f the motions associated
b o d y .177 Michael Scot offered very nearly the same argument.178
with the latter. John o f Sacrobosco, for example, says no more about the
A t some point, however, perhaps during the initial influx o f Greco-
ninth sphere than that it is the “ first mobile orb” (primutn mobile), or the
Arabic astronomical and cosmological literature, the starless, transparent
“ last heaven.” 174 Other thirteenth- and fourteenth-century authors - for
ninth orb was linked with the biblical waters above the firmament and
example, Robertus Anglicus and Cecco d’Ascoli - also mention it without
thereafter identified with the crystalline orb. The two are already joined in
any reference to its “ crystalline” character or its theological connection.
Vincent o f Beauvais’ Speculum naturale, probably composed over the period
Precisely who linked the biblical account with a ninth and starless orb
1244-1254.1 9 Although, as we saw, not everyone explicitly made the con­
may never be known. As early as the sixth century, John Philoponus iden­
nection when discussing the ninth orb, many did, so that throughout the
tified the ninth sphere,175 not with the waters above the heaven, but with
period o f this study the ninth sphere was frequently equated with a ervs-
the heaven created on the first day, which he described as a transparent,
talline orb o f biblical origin.
starless orb that surrounds the firmament created on the second day.176 But
We have now considered all the mobile orbs in medieval cosmology but
quae super coelos sunt, laudent nomen Domini. Neque incongrue hi duo orbes aquarum have yet to examine the final theological sphere, the immobile, empvrean
nomine designari possunt propterea quod cum nullae in iis stellae fulgeant, sed admodum heaven. Because the latter was widely discussed and always problematic, a
translucidi et perspicui sint aquarum referunt similitudinem turn quia pertrigerandi vim
habere creduntur. Quare a quibusdam coelum aqueum, sive glaciale, ab aliis Christal-
separate chapter (Ch. 15) will be devoted to it.
linum uno communi nomine appellantur.” Commbricenses, De coelo, bk. 2, ch. 5, qu.
177. Robertus Anglicus. Sphere, lec. 1, 1949, 148 (Latin), 203 (English).
1. art. 5, 1598, 252.
170. He was uncertain in the Theorica planetarum but not in Tracuuus de sphera. See Campanus 178. Michael’s discussion is embedded in a question format. See Michael Scot, Sphere, lec.
2, 1949, 259-260.
of Novara, Theorica planetarum, 1971. 183 and 393, n. 53.
171. “ Propter secundum sciendum quod theologizantes plures speras, sive celos, figurant, 179. In his lengthy discussion on the crystalline sphere (bk. 3, chs. 90-100. cols. 221-229;.
super octavam, quidem, ponentes cristallinum sive aqueum, et empireum seu igneum, Vincent focuses mostly on its theological aspects, but in chapters 97 and 100. columns
de quibus in presentanum nihil, cum potius que ipsorum, revelatione quam ratione 226 and 228, respectively, he links it explicitly with an astronomical ninth heaven.
sciantur, aut experientia, quibus hoc opus nititur sistere. ” Peter ot Abano, Lucidator.
differ. 3, 1988, 214, lines 4-8.
172. Nowhere in the works o f Aristotle is there even a vague hint o f a ninth orb beyond the
fixed stars.
173. Bacon, De celestibus, pt. 4, ch. 2, Opera, fasc. 4, 1913, 387-388.
174. See Sacrobosco, Sphere, ch. 1, 1949* 118—119.
175. He attributed the discovery o f a ninth sphere to Hipparchus and Ptolemy. In the Hy­
potheses, Ptolemv seems to have assumed a ninth sphere, whose functi* m was to move
the starry sphere but which was not connected with the precession os the equinoxes.
See Ptolemy, Hypotheses, 1907, 122, 125, and Pedersen, 1974, 249, n. 7.
176. For a summary o f Philoponus's views in his De opijicio mundi. see Duhem, Le Systeme,
1913—1959* 2:496-501.
ARE THE HEAVENS H ARD OR FLUID? 325
14 possible. Thus to inquire about the possible hardness or softness o f celestial
orbs is to ask an irrelevant question.
Perhaps in order to avoid posing such questions, William Donahue sup­
poses that the Peripatetics conceived o f the heavens and its orbs as “ im­
Are the heavens composed of material or quasi-material” entities, whatever these terms may signify/ A
similar position is adopted by Edward Rosen, who denies (1985, 13) that
hard orbs or a fluid Aristotle argued for the existence o f solid celestial orbs. Because Aristotle
denied the existence o f corruptible terrestrial matter in the incorruptible and
substance? unchanging ethereal heavens, it followed, Rosen argued, that his celestial
ether was not material and therefore could be neither solid nor hard, as
Pierre Duhem had claimed. Indeed, by similar reasoning (though Rosen
does not draw the inference), it could not be fluid either. From a narrow,
strict-constructionist standpoint, one might defend such an approach. In
reality, however, this interpretation misunderstands Aristotle’s intent, and,
Prior to the impact o f Tycho Brahe’s astronomical research, scholastic au­
if applied to the Middle Ages, would distort medieval opinion. It will be
thors found no reasop to devote even a single question to consider whether
well to eliminate this second interpretation before proceeding.
the celestial orbs might be hard or soft. Tycho, however, had made the
Although Aristotle may have denied that alterable matter like that on
question virtually unavoidable. The issues he raised challenged the very
earth could exist in the heavens, his ether may be construed as a fifth kind
existence o f eccentric and epicyclic orbs and inevitably posed questions
o f substance, or element - a quinta essentia, as many commentators would
about the hardness or softness o f the celestial ether. Although it was Tycho
call it — with properties, as we have seen, radically different from those o f
who first made the ether’s hardness or softness an issue central to cosmology,
the four sublunar elements. Whatever Aristotle may have thought about
the problem had a long, but vague and even muted, history. Because me­
the properties o f the celestial ether, there is no doubt that in De caelo he
dieval scholastic natural philosophers rarely discussed the matter directly or
assumed the corporeality, and therefore phvsicalitv, o f the heavenly orbs.3
in useful detail, information about the hardness or fluidity o f orbs must be
As nonspiritual, corporeal, and therefore three-dimensional physical entities
gleaned indirectly from discussions in other contexts.
composed o f ether, celestial orbs had to be something akin to hard or soft
- even though Aristotle himself was committed to a formal denial o f con­
trary qualities such as hardness and softness, hotness and coldness, and rarity
I. M odern in terp retation s o f m ed ieva l orbs and density. In the course o f discussions on the celestial orbs, one would
sooner or later find it necessary to speak o f their physical nature, despite
A widely held opinion today is that scholastic authors thought the celestial Aristotle’s strictures. Were they hard or soft? If one or the other, then could
orbs were solid, where “ solid” is taken as synonymous with hard or rigid.' they also be said to be, in some sense, dense or rare? But Aristotle seems
Here the image is one o f transparent glass or crystalline globes. Hardly in to have precluded such analyses. Perhaps this is w hy he chose to ignore the
contention as to popularity with the first opinion today is a second, which physical nature o f celestial spheres and w hy he offered no helpful clues as
assumes that medieval thinkers faithfully adhered to Aristotle’s dicta about to how one might speak about them.
the celestial ether. Thus the orbs or spheres could be neither solid nor fluid Indeed, this may well explain w hy his medieval scholastic commentators
because Aristotle had argued that contrary qualities such as hardness and also chose to ignore the problem. Butjust as many scholastic authors ignored
softness, density and rarity, and so on, were inapplicable to the incorrup­
tible, celestial ether o f which they were composed. Nicholas Jardine ob­ 2. Sec Donahue, 1973, 251. 256-259, 275.
3- In De caelo 2.12.2933.8, Aristotle declares that “ the last sphere moves round embedded in
serves (1982, 175) that to pose a question about the hardness or softness ot a number ot'spheres, and each sphere is corporeal." Aristotle [Guthrie], i960. In the Latin
celestial spheres would have been considered a “ category mistake.” Hard­ translation o f Aristotle’s De caelo that accompanies Averroes’ commentary, the Latin trans­
ness and softness are qualitative opposites found only in terrestrial matter. lation o f Aristotle's second phrase is “ et omnis orbis eorum est corpus.” See Averroes [De
caelo, bk. 2, text 70], 1562—1574, 5:70r, col. 1. In his Hexaemeron, Robert Grosseteste
Since pairs o f opposite qualities are the source o f all terrestrial change, they reports: “John Damascene also implies in his book o f Sentences that the existence o f an
must o f necessity be absent from the celestial region, where change is im-1 immaterial body, that which is called a fifth body among the wise men o f the Greeks, is
impossible," but he declares that Aristotle and his followers did assume the existence o f
a “ fifth body” in addition to the four elements. Grosseteste, Hexaemeron. part. 3, ch. 6,
1. For a lengthy list ot" scholars who hold this opinion, see Grant, 1987a, on which I rely tor 1. 1982, 106. John may have had in mind the passage from Aristotle just cited.
much o f what follows. See 153.
326 THE C E L EST IA L REGION ARE THE H E A V E N S H A R D OR FLU ID? 327

Aristotle’s famous dictum that neither place, nor void, nor time could exist agreement about the nature ot the different traditional divisions o f the heav­
beyond the outermost sphere ot the physical world and began to inquire ens, which often influenced the properties that were assigned. By the
what indeed might lie beyond, so, to a lesser extent, did some o f those seventeenth century, Giovanni Baptista Riccioli distinguished five different
same scholastic authors reveal an opinion or judgment, usually indirectly, interpretations concerning the hardness or softness o f the heavens.5 There
about the hardness or softness ot the celestial orbs, which they all assumed were those who believed that all the heavens were w holly solid - that is,
to be physical bodies. hard - while others thought them w holly fluid. The last three opinions
To my knowledge, no medieval natural philosopher rested content to assumed a partly solid and partly fluid celestial region. Thus Riccioli reports
depict the celestial orbs as immaterial entities devoid o f physical properties. a third opinion wherein the fluid part apparently consisted o f ring-like
This remains true, despite the tact that many denied in the abstract that the channels filled with a subtle or tenuous air-like substance. The surfaces and
celestial ether could possess terrestrial attributes such as hotness and cold­ everything else were presumably hard. A fourth opinion assumed that the
ness, or rarity and density. O nly when confronted with specific problems extremities o f the celestial region were solid, namely the heaven, or sphere,
about the spheres themselves - that is, about their arrangement and the o f the fixed stars and the heaven, or sphere, o f the Moon. Everything
relationships between successive surfaces — do scholastic natural philoso­ between these two extremes was o f a fluid nature. The fifth and final opin­
phers speak in a quite different vein and reveal, perhaps inadvertently, a ion, which Riccioli says is “ now the most celebrated” (nunc celeberrima
concern about real, physical spheres. Indeed, we have already seen that opinio) - indeed it wras a direct legacy from Tycho Brahe - assumed a solid
numerous medieval discussions about possible physical problems that might sphere for the fixed stars with the planetary heavens being o f a fluid nature.
affect eccentric orbs — for example, whether vacua can occur between suc­ A m ong the numerous partisans o f these opinions that Riccioli mentions,
cessive celestial surfaces or whether two orbs can overlap and occupy the few are from the late Middle A ges.6 Indeed, Riccioli includes only the names
same place simultaneously - provide ample evidence that the spheres were o f Thomas Aquinas, who, he says, took no position on the fluidity or
conceived as physical bodies. I am aware o f no instance in which physical solidity o f the heavens; Michael Scot and Cecco d’Ascoli, who assumed
considerations were dismissed because the celestial orbs were deemed im­ that the heavens were fluid; and Saint Bonaventure, who assumed that the
material or quasi-material. Despite Aristotle’s well-known attitude, those heavens were partly fluid, partly solid.7 Although, as we shall see, other
orbs were judged to be physical, and it was therefore difficult to avoid the names may be added, the paucity o f medieval names is probably no ov­
attribution o f some physical properties to them. Although the attribution ersight but reflects the fact that tew medieval scholars expressed opinions
o f contrary qualities to Aristotle’s ethereal orbs is a “ category mistake” on this interesting problem, which only became a major issue in the late
w'ithin Aristotelian cosmology, some o f Aristotle’s legions o f commentators sixteenth and the seventeenth century. It was during this later period that
often found it unavoidable to attribute terrestrial qualities to celestial bodies. the expression “ solid sphere” became virtually synonymous with “ hard
With the second interpretation eliminated from further consideration, we sphere” or “ rigid sphere.” From the seventeenth century on, the association
shall now attempt to determine whether, during the late Middle Ages, the o f “ solid” with “ hard” was applied retrospectively to the Middle Ages and
celestial orbs were conceived as hard and rigid or fluid and soft, or perhaps became fixed in the subsequent literature o f the history o f astronomy and
some combination o f these properties. The problem o f the hardness or cosm ology. Thus it is that when modern scholars speak o f “ solid” orbs in
softness o f the celestial region is rather complex. The fact that a natural the Middle Ages, they usually mean orbs that are hard and rigid.8 But was
philosopher may have assumed the existence o f celestial orbs does not permit this the medieval understanding o f a celestial orb? To ascertain whether the
us to make any inference as to whether he thought them hard o: soft. description o f an orb as solid also implied its hardness or rigidity, we must
Ptolemy himself may have aided the confusion by his apparent assumption
in the Almagest that planets move about in fluid media, a view he seems to 5. See Riccioli, Almagestum novum. pars post., bk. 9, sec. 1, ch. 7, 1651, 238, col. 2-240, col.
have abandoned in the Hypotheses oj the Planets.* M oreover, there was dis-4 *
6. Most are from the ancient period (Greeks and Romans) and from the sixteenth and sev­
enteenth centuries, which would have been the modern period for Riccioli.
4. Duhem, Le Systeme. 1913—1959, 2:479, explains chat in the Almagest Ptolemy regarded the 7. Riccioli, Almagestum novum, pars post., bk. 9, sec. 1, ch. 7, 1651, 239, cols. 1—2. Bonav-
heaven as a fluid in which the stars moved freely but that he abandoned this idea in his enture’s position is unclear. Although he accepts the existence o f orbs, he also insists that
later Hypotheses. In his translation, Talialerro has Ptolemy say that the planets “ can all they are continuous, subtle, and rare like water. Indeed, they have no terminating surfaces
penetrate and shine through absolutely all the fluid media” (Ptolemy, Almagest, bk. 13, as do solid bodies (see Bonaventure [Sewena’s. bk. 2, dist. 14, pt. 2, art. 1, qu. rj, Opera,
ch. 2 [Taliaferro], 1952, 429). in Toomer’s translation (Ptolemy. Almagest, bk. 13, ch. 2, 2:352, cols. 1-2). Bonaventure’s heavens seem more fluid than solid. Nevertheless, Amicus
1984, 601), however, Ptolemy says that the nature o f the heavens “ is such as to afford no also attributed to Bonaventure heavens that were a mixture ot fluid and solid (Amicus [De
hindrance, but o f a kind to yield and give way to the natural motions o f each part, even caelo, tract. 5, qu. 5. art. 2], 1626, 275, col. 1).
if [the motions] are opposed to one another.” Is a substance that yields and gives way to 8. For example, Jardine, 1982. 175, assumes that solidity is the opposite o f fluidity and is
the motions of other parts also a fluid substance or merely elastic? therefore equivalent to hardness.
328 THE CELESTIAL REGION ARE THE HEAVENS HARD OR FL U ID ? 32 9

examine the meanings that were assigned to the term solidum during the [i.e., the convex surface] that a round solid is contained, because it includes every­
Middle Ages.

II. T h e m ean in g o f the term so lid u m in the M id d le A ges


|
j
thing within itself, leaving nothing outside.

Is there a parallel between the respective threefold senses o f the terms "solid”
and "surface” ? Since Michael obviously intended three senses o f the term
At the beginning o f his famous thirteenth-century treatise On the Sphere 1 surface, may we also infer that he intended three distinct senses o f the term
(De spera), John o f Sacrobosco defined a sphere by citing Euclid and Theo­ | solid? If so, then he may also have intended that celestial bodies be conceived
dosius, both o f whom considered it a solid body.910As a consequence, it j of as continuous but not hard, or at least not necessarily hard.
became usual, in commentaries on Sacrobosco’s Sphere, to inquire about An examination o f a discussion by Robertus Anglicus, in the latter’s
the nature o f a sphere and occasionally to ask about the sense in which a I Commentary on the “ Sphere” of Sacrobosco, written around 1271, lends support
sphere was a solid. In a commentary on the Sphere ascribed to Michael Scot, ] to this possibility. In a passage that he may have drawn, and perhaps even
we learn that the term solidum is spoken o f in three ways: 1 copied, from Michael Scot, Robertus describes the same three senses o f the
« terms solidum and superficies. 12 He assumes the existence o f nine celestial
in one way, it is the same as hard, just like earth; in another w ay solid is the same orbs and also proclaims the immutability o f the material from which they
as continuous, and thfhs the elements and supercelestial bodies are called solid; in a are composed. The orbs are distinguished as being larger and smaller orbs
third way it is like a three-dimensional thing, and thus it is the same as a bodv. by means o f "greater and smaller intelligible [i.e., imaginary] circles.” That
Therefore it is not superfluous to say that a s p h e r e is a s o l i d b o d y .'0 is, an orb, or sphere, is the space that is cut o ff between two such circles
and is the place where each planet is carried.'3 Robertus illustrates the ar­
Although this significant passage poses serious problems, it is striking that rangement o f the nine celestial orbs by imagining nine wheels o f such sizes
Michael - for convenience, let us assume that he was the author — invokes that they can be arranged concentrically. These nine nested wheels are
earth to exemplify the meaning o f a hard solid, but mentions celestial bodies" assumed to be in the air and to move around the same center. Robertus
(and the elements) to illustrate the meaning o f a continuous solid. Does this now explains that the quantity or volume o f air between any two wheels
signify that Michael thought o f the celestial bodies as continuous but not is like a celestial orb which carries around the planet that lies within it. By
hard? This may depend on whether the term ‘‘elements’’ (elementa), in the choosing air enclosed by wheels as his analogy with celestial orbs, Robertus
second sense o f solid, includes or excludes earth. Was Michael, in effect, leaves the impression that he conceived o f the celestial orbs as somehow
dividing the elements into hard (earth) and soft (water, air, and fire), with fluid in nature - or at least fluid in their interiors, if not in their surfaces.
only the latter assumed continuous? If so, the celestial bodies would also This interpretation gains credibility when Robertus later considers
be continuous and soft, just like water, air, and fire. whether the celestial spheres are continuous (continue) or contiguous (con-
Another possibility suggests that Michael had something else in mind, tigue) and decides that they are continuous, which means that the convex
namely to signify that solid bodies possessed all three attributes: hardness, surface o f one sphere is identical with the concave surface o f the next-
continuity, and tridimensionality. This interpretation seems less plausible, superior o rb .'4 But if the successive surfaces o f successive orbs are contin­
because, in a sentence immediately following the one proclaiming his three­ uous, Robertus acknowledges the following problem:
fold sense o f the term solidum, Michael provides a clue that he may have
intended three quite distinct senses o f solid rather than to suggest that a since the orbs are moved by contrary motions. . . . then one and the same [surface]
solid possesses all three attributes. For he says: would be moved by contrary motions, which is impossible. Also, it would then
follow that, if one orb were moved by some motion, all the other orbs would be
It is also known that a surface [supe rf ici es ] is threefold: it is plane, as in a wall: it IS moved by the same motion, which, nevertheless, we know to be im possible.'5
concave, as in a tub; [and it is] convex, as in a mountain. And it is by such a surface
12. For the Latin passages, see Robertus Anglicus. Sphere, lec. 1, 1949, 145 (and 200 for
9. See Sacrobosco, Sphere. 1949, 76—77 (Latin) and 118 (English). Thorndike’s translation). Robertus substitutes the expression corpora celestia for Michael
10. “ Item nota quod solidum dicitur tribus modis: uno modo est idem quod durum, sicut Scot’s corpora supercelestia. Aiton, 1981, 90, mentions Robertus’s discussion o f the term
terra; alio modo solidum idem quod continuum, et sic elementa et corpora supercelestu “ solid” and correctly explains that “ The Earth was solid in the first sense, while the
solida dicuntur; tertio modo. id est, quod trina dimensio. et sic idem est quou corpus, celestial bodies were solid in the second and third senses, but were not necessarily hard.”
unde non est ibi negatio, Spera est corpus solidum." Ibid., 256. r3- Robertus uses the terms spera and orbis interchangeably.
11. By the expression corpora supercelestia, I assume that Michael means all celestial bodies J4 - “ Ad primam questionem dicendum quod omnes orbes novem sunt continui.” Robertus
that is, both planets and spheres. Anglicus, Sphere, lec. 1, 1949, 147. If the surfaces were distinct, the orbs would be
contiguous. This discussion appears in ibid., 146-147 (Latin), 202-203 (English).
! r5- Ibid., 202. These arguments have already been described.
330 THE C E LESTIA L REGION ARE THE HE A V E N S H A RD OR FLUID? 33 1

In replying to this difficulty, Robertus indicates that orbs are fluid when he may be drawn from the discussions o f our two thirteenth-century com­
says: mentators on the Sphere o f Sacrobosco: when the term solidum, or any o f
its variants, occur in the context o f a discussion on celestial orbs, they may,
We suppose the outer edge o f the orb im m obile and the middle o f the orb to be in the absence o f any other decisive criteria, refer to either hard or soft
moved, just as we see that the center o f water is moved, yet at its sides the water spheres. The modern interpretation, which always equates solid with hard,
is still. And it seems much more likely that this can be done in the orbs, which are is unwarranted.
much simpler than water. N or, as is now clear, need all orbs be m oved when one But a m ove toward equating solid with hard may already have been
orb moves, although they are continuous, just as it is not necessary that, when a under w ay in the fourteenth century. We may infer this from Pierre d’Ailly,
part o f the water is moved, all the water should be moved, although the water is who, by the end o f the fourteenth or early fifteenth century, said that solidum
continuous.1'' is taken in three ways, the first o f which assumes that it is “ firm or hard,
just like iron or stone, and this is the common u s a g e . I n the course o f
Whatever Robertus may have meant by these examples, the fact that he the sixteenth century, and certainly by the seventeenth century, the earlier
used water to illustrate the continuity and motion o f celestial orbs suggests ambivalence vanishes: fluid then was regularly opposed to solid, with the
that he thought o f those orbs as continuous and fluid rather than as con­ latter unequivocally equated with hard, as when Johannes Poncius declared:
tinuous and hard.'7 Moreover, his description is o f great interest, for he “ some say that the heaven is a continuous body and fluid, like a ir. . . ; and
seems to say that the surface o f an orb can be assumed to be immobile while other moderns think that the firmament is a solid [i.e., hard] b ody.” "'' For
the part toward its center is in motion. Thus the planet itself is somehow Michael Scot and Robertus Anglicus, and perhaps tor other natural philos­
carried by the fluid part o f the orb lying within its immobile surfaces. But ophers during the thirteenth to fifteenth centuries, the concept o f a solid
what is the nature o f an orb’s immobile surfaces? The water analogy indicates body may have entailed a fluid state. But during that very period the iden­
fluid surfaces, since the latter are in no w ay differentiated from their mobile tification o f hard with solid was well under way.
content.
By adopting an approach in w’hich only the inner part o f an orb was
assumed to move while its outermost surface lay immobile, Robertus
avoided the seemingly impossible dilemma that would have resulted from III. T h e three m ajor positions
an assumption o f the continuity o f the celestial orbs. For on that assumption, From m y summary o f Riccioli’s five opinions on the fluidity or hardness
two separate but successive and immediately proximate orbs would nec­ o f the heavens, it is evident that these opinions reduce to three alternatives:
essarily move in the same direction, because the convex surface o f the inner (1) all are hard; (2) all are fluid, or soft; and (3) some are hard and some
sphere would be continuous - that is, identical - with the concave surface fluid.'0 M ost scholastics appear to have adopted the third, or mixed, inter­
ol the next-superior orb. Despite the assumption o f continuity, however, pretation, assuming some orbs to be hard and others soft. A prime example
Robertus could now declare that although the convex surface o f the pla­ is Aegidius Romanus, who assumed a hard overall orb from the Moon to
netary orb ot Mars and the concave surface o f the sphere o f Jupiter, for the sphere o f the fixed stars but conceived o f the eccentric deferents as
example, wrere one and the same, those tw o planetary spheres could, none­ channels filled with a soft substance. Thus far the authors whom we have
theless, move in different directions, because only the middling parts of sampled have spoken only in generalities. We must now examine medieval
each sphere or orb - not the surface itself - would move.
Although the evidence is stronger for Robertus Anglicus, both he and
18. “ Esc advertendum quod tnpliciter solet capi solidum. Primo modo prout tantum valet
Michael Scot appear to have thought o f the celestial orbs as soft rather than sicut firmum vel durum sicut terrum vel lapis et sic eo utuntur vulgares.” D ’Ailly [14
hard. Whatever the merits o f that claim, however, one strong inference16 *
7 Questions, qu.i], 1531, rqSr.
19. Poncius wrote these remarks in his commentary on book 2 of Scotus’s Sentences. See
Poncius [Sentences, bk. 2, dist. 14, qu. 2] 1639, vol. 6. pt. 2:727, col. 1. Poncius observes
16. Ibid.. 202-203. that the solidity - that is. hardness - o f the heavens is an opinion "more common" to
17. Bonaventure seems to have held the same opinion (see note 7 o f this chapter). Aegidius Peripatetics (727. col. 2). Although Amicus accepted solid and hard celestial orbs, he
Romanus may have had Robertus in mind when he attacked, as a “ fatuous” opinion, the reports that “ in bk. 1, ch. 1 o f his Almagest, it seems that Ptolemy assumes that solid and
idea that "just as water that remains continuous can be moved according to one part to hard bodies are mutually distinct" (Videtur Ptolomaei I, Almeg. c. 1 . . . supponit esse
one place and according to mother part to another place: thus the orbs should be able to corpora solida et dura inter se disuncta). See Amicus, De caelo. tract. 5. qu. 5, art. 2.
do the same thing. But thf> cannot occur without the division o f the water. Hence this 1626, 275, col. 2.
cannot be applied to the orbs without division o f the orbs." Aegidius Romanus, Opus 20. Among authors who distinguish these three positions are Amicus, ibid., 275, col. 1 and
Hexaemeron, pt. 2, ch. 32, 1ss>, 49r. col. 2. Aegidius regarded the heavens as solid overall, Roderigo de Arriaga [De caelo, disp. 1, sec. 3, subsec. 1], 1632. 499, cols. 1-2. Both
with softness confined to the channels marking out the eccentric deferents. present four different opinions.
332 THE CELESTIAL REGION ARE THE H EAVENS H A R D OR FLUID? 333

opinions about the orbs or heavens themselves. Because so few considered declares that the waters above the heaven (i.e., firmament) are not in a
the various parts o f the heavens systematically, information must be gleaned vapory form but are suspended by icy solidity (glaciali soliditate) to prevent
from a variety o f sources, often from discussions o f this or that particular their fall.234 Here we find the conviction that in liquid form the waters would
12
heaven or o f the heavens as a whole. Therefore, I shall now describe what surely flow downward; only if they were frozen or congealed would they
medieval natural philosophers thought about the hardness or softness o f the remain suspended above the firmament.
tw o major celestial subdivisions, namely the crystalline heaven and the Those who assumed that the waters above the firmament were fluid
firmament, the latter, as we saw (Ch. 5, Sec. VI. 2), often including the formed the larger group during the Middle Ages and included Ambrose,
planetary orbs, which were frequently conceived as subdivisions of John Damascene, Alexander o f Hales, Robert Grosseteste, Richard o f Mid­
the firmament. dleton, Saint Bonaventure, Vincent o f Beauvais, and an anonymous author
o f a French encyclopedia written around 1 4 0 0 . Within this group, some,
like Richard o f Middleton and Saint Bonaventure, provide little or no de­
scription o f the other alternative, namely that the supraheavenlv waters
IV . T h e crysta llin e orb might be hard like a crystal. They were agreed, however, that although
In its theological aspect, the crystalline sphere developed from commentaries these waters were unlike ordinary elemental water, they did share with it
on Genesis 1.7, which spoke o f a division o f waters between those above a few important properties, namely, transparency (perspicuitas) and coldness
the firmament and those below. From the time o f the Church Fathers, the (jrigiditas), and for Richard also the property o f wetness (humidum). Vincent
meaning and significance o f the waters above the firmament were, as we o f Beauvais, who assumed that the waters were immutable, believed they
have seen, much debated. Because the biblical text spoke o f waters above were luminous (luminosum), transparent (perspicuum), and subtle (subtile).2A
the firmament, Christian authors, following Saint Augustine, were gen­ Although terms like “ crystalline” and “ icy solidity” seem to imply hard­
erally agreed on the necessity for a literal interpretation o f this particular ness, they could be interpreted otherwise. In commenting on Peter Lom­
text and were therefore committed to the existence o f waters o f some kind bard’s passage on the icy solidity o f the waters above the firmament,
above the firmament (see Ch. 5, Sec. V I).'1 All else was seemingly arguable. Bonaventure insists that the sense o f solidity that implies that those waters
Indeed, the debate hinged on the interpretation placed on the terms “ waters” are heavy and held in position above the firmament by violence is contrary
(aquae) and “ firmament” (firmamentum), the latter largely determining the to the order o f the universe. We should rather understand that “ those waters
meaning o f the form er.'' compare with icy solidity [glaciali soliditati] not because o f heaviness but
From the time o f the Church Fathers to the end o f the Middle Ages, a because o f continuity and stability because they do not ebb or flow, nor do
variety o f interpretations o f the waters above the firmament were proposed. they descend dow nw ard.” 27 Bartholomew the Englishman is even more
The interpreters divide essentially into two groups: those who thought of explicit when he explains that the waters above the heaven (firmament) are
the waters as solid and hard and those who considered them fluid. Among called “ crystalline, not because they are hard [durum] like a crystal but
the former we may include Saint Jerome and Bede, the latter likening the
24. See Peter Lombard [Sentences, bk. 2. Hist. 14, ch. 4, par. 1], 1971, 396.
waters to “ the firmitv o f a crystalline stone” (cristallini lapidis quanta jtrrtn- 25. Saint Ambrose, Hexameron. 3rd homily (bk. 2. the 2nd day), ch. 3, 1961. 51-59; John
tas).ZJ In his famous and widely used Sentences, composed in the twelfth Damascene, The Orthodox Faith, bk. 2, ch. 9 [Chase], 1958, 37:224; Alexander ot Hales,
Summa theologica. mquis. 3, tract. 2, qu. 2, tit. 2, ch. 5 (“ Qualiter lllae aquae dicantur
century, Peter Lombard indicates an awareness o f Bede’s opinion when he
caelum crystallinum” ), 1928. 2:341; Grosseteste, Hexaemeron. part. 3, ch. 5, 4. 1982, 104;
Richard o f Middleton [Sentences, bk. 2. dist. 14, qu. 1 (“ Utrum coelum crystallinum
21. One who was not was William o f Conches (fl. 1120-1149), who. in his Philosophia mundi. dictum sit naturae aquae")], 1591, 2:167-168; Bonaventure, Sentences, bk. 2, dist. 14, art.
denied that waters could exist above the firmament and insisted that the scriptural passage 1, qu. 1 (“ Utrum caelum crystallinum sit natura aquae” ). Opera. 2:335—338; Vincent ot
in which this is asserted must be interpreted allegorically. See Lemay. 1977, 231. Beauvais, Speculum naturale, chs. 90—roo, 1624, cols. 221—229, especially col. 224; tor the
22. To see how the meaning o f “ firmament” determined the meaning o f the “ waters” above statement in the French encyclopedia, see Hvatte and Ponchard-Hyatte, 1985, n .
that firmament, and to obtain an excellent sampling o f the different interpretations placed 26. Richard is silent, whereas Bonaventure mentions only that Bede believed that the waters
on both o f these terms, see Thomas Aquinas. Summa theologiae, pt. 1, qu. 68, arts, i above the heavens rested, and were sustained, by virtue o f their solidity: "Et ibidem
(“ Was the firmament made on the second day?” ); 2 (“ Are there any waters above the aquae illae quiescunt et sustentantur vel sua soliditate. sicut videtur Beda dicere, vel sua
firmament?” ); and 3 (“ Does the firmament separate some waters from others?” ), I964_ subtilitate, vel etiam Dei virtute, quae sic ordinavit. ’ Bonaventure, ibid., 337, col. 2. For
1976, 10:71-77. 77-83, and 83—87, respectively. Aquinas explains (79): “ we maintain that Vincent, see Speculum naturale. bk. 3, ch. 95, 1624, i:col. 224. Vincent also describes the
these waters are material. Just what they are must be explained in different ways depending alternative opinion that the waters are congealed like a crystal and rejects it because he
on various theories about the firmament.” He then proceeds to offer a number ot inter­ can find no cause that would congeal the waters; ibid., col. 221. Grosseteste described
pretations (for some o f them, see Ch. 5, Sec. VI). the opinion that the waters above the firmament were like a hard, crystalline stone (cristallus
23. For the relevant passages from Jerome and Bede, see Campanus of Novara, Theortca lapis) and also rejected it (Hexaemeron, part. 3, ch. 3, 4, 1982. 104).
planetarum, 1971. 393—394, n. 54. 27. Bonaventure, Sentences, bk. 2, dist. 14, pt. 1, art. 3, qu. 2, dub. 1, Opera, 2:350, col. 1.
334 THE CE LEST IA L REGION ARE THE H E A V E N S HARD OR FLUID? 335

because they are uniformly luminous and transparent. Moreover, it [i. e., unusual for natural philosophers to extend the firmament to include not
the crystalline heaven] is called watery insofar as water is moved by virtue only the eighth sphere o f the fixed stars but also the planetary orbs. N ow
o f its subtlety and mobility. ” ‘ s A similar description is provided by the we must determine whether that firmament was conceived as hard, or fluid,
author o f a French encyclopedia, who, around 1400, declared that or some combination o f these contrary properties
Arriving at an opinion was not often easy, as two o f the greatest Church
Others call it [the ninth sphere] the “ crystalline sphere” or the “ crystalline heaven” Fathers, Augustine and Basil, illustrate. Despite his observation (in his com­
[or sky], not because it is o f hard and solid material like crystal, but for its luminosity mentary on Genesis), that too much subtlety and learning had been ex­
and its great transparency and uniformity. And it is also the heaven [or sky] that pended on explicating the nature o f the firmament,31* Augustine advised
theologians call “ w atery,” not because there are waters such as those which are here those who analyzed the meaning ot the firmament to “ bear in mind that
below, rather they are light [s o u b ti ll e s ] waters o f a noble nature similar to the heaven the term ‘firmament’ does not compel us to imagine a stationary heaven:
[or sky] in clarity and luminosity/'' we may understand this name as given to indicate not that it is motionless
but that it is solid and that it constitutes an impassable boundary between
Vincent o f Beauvais declares that the waters above the firmament should the waters above and the waters b elow .” 5* Without choosing between them,
not be understood as the element water but rather as a kind o f diffuse matter Augustine thus explained the “ firm ity” o f the firmament in two ways: it
(materia confusion::) that has within itself “ hotness, dryness, wetness, and is firm either because it is motionless or because it is solid and prevents the
coldness, luminosity, darkness, transparency, and opacity.” 2 30 Hardness is
9
8 passage o f waters from above or below. He gives no indication that those
not even mentioned. who assumed a “ solid” and impenetrable firmament meant also to signify
For many, if not most, ot those who considered the suprafirmamental that it was hard. However, elsewhere in the same commentary on Genesis,
waters “ crystalline,” the latter term did not signify the hardness o f the Augustine speaks o f air and fire as material constitutents ot the heavens,
waters but rather their immutability, transparency and luminosity. When thereby suggesting soft and fluid heavens.33*
medieval authors spoke o f the crystalline sphere, they usually had in mind Saint Basil suffered from similar equivocation, which he exhibited over
those properties o f a crystal such as luminosity, transparency, and even a the span o f a few lines o f his Hexaemeron, where he wrote:
quasi immutability, rather than hardness.
N ot a firm and solid nature, which has weight and resistance, it is not this that the
word “ firmament” means. In that case the earth would more legitimately be con­
sidered deserving o f such a name. But, because the nature ot superincumbent sub­
V . T h e firm am en t and the p la n etary orbs
stances is light and rare and imperceptible. He called this firmament, in comparison
In Chapter $, we saw the variety o f meanings that attached to the term with those very light substances which are incapable o f perception by the senses.14
“ firmament” (firmamentum), the heaven that God created on the second day
(Genesis 1.6-8) to divide the waters above from the waters below and in With his denial o f solidity to the firmament, Basil goes on to deny as well
which he placed the luminaries on the fourth day (Genesis 1.14—19) to divide that it could be composed o f the simple elements or o f any combination o f
day and night. Because o f the explicit biblical assertion that the luminaries them .35 Despite his uncertainty about its composition, Basil seems to incline
were in the firmament (“ Fiant luminaria in firmamento caeli” ), it was not toward a fluid firmament.
28. “ Et ideo in summo dicitur crystallinum non quia durum sicut crvstallus, sed quia uni- 31. Augustine proclaimed that he himself had “ no further time to go into these questions
tormiter est luminosum et perspicuum. Aqueum autem dicitur quemadmodum aqua ex and discuss them, nor should they have time whom I wish to see instructed for their own
sua subtilitate et mobilitate movetur. ” Bartholomew the Englishman, De rerum proprie- salvation and for what is necessary and useful in the Church.” Augustine, Genesis, bk.
tatihus, bk. 8, ch. 3 (“ De coelo aqueo sive crystallino"), 1601, 379. As we saw, Vincent 2, ch. 8, 1982, 1:60-01.
o f Beauvais held a similar view about the crystalline sphere. 32. Ibid.
29. Hyatte and Ponsard-Hvatte, 1985, 11. I am grateful to Prof. Reginald Hvatte. who not 33. See Chapter 10, note 4.
only called mv attention to the reference but also translated the relevant passage. 34. Basil, Exeqetic Homilies (Hexaemeron), homily 3, 1963, 47. Earlier m homily 3 (p. 43),
30. “ Nos itaque dicimus quod ubi dicitur Deus divisit per firmamentum aquas ah aquis, non Basii had denied that the firmament could be compared to water that is "like either frozen
sumitur aqua pro elemento, sed pro materia confusionts, quae habet in se calidum et water or some such material which takes its origin from the percolation o f moisture, such
siccum, humidum et frigidum, luminosum et tenebrosum, perspicuum et opacum.” Vin­ as is the crystalline rock which men say is remade by the excessive coagulation o f the
cent of Beauvais, Speculum na>urale, bk. 3, ch. 95, 1624, i:col. 224. Later, in ch. 95, col. water, or as is the element o f mica which is formed in mines.”
225, Vincent explains that “ this heaven [the crystalline] is not crvstalline with respect to 35. See Chapter 10, note 4. But in homily 3, Basil seems to equate the heavens with the
every property o f a crystal, but according to the property o f transparency." (Dicimus firmament when he declares (ibid., 49-50): “ w'e have observed in many places that the
quod coelum lllud non est chrystallinum secundum omnem propnetatem chrvstalli. sed visible region is called the heavens due to the density and continuity o f the air which
secundum propnetatem perspicui.) clearly comes within our vision and which has a claim to the name of heavens from the
336 THE CELESTIAL REGION ARE THE HEAVENS HARD OR FLU ID? 337

A solid, hard firmament had at least one important defender in late an­ During the late Middle Ages, most authors were vague and noncommital,
tiquity. Approximately tw o hundred years after Saint Basil, John Philo- despite the fact that the very name firmamentum, with its implications o f
ponus considered the nature o f the firmament in his commentary on the strength, power, stability, and even o f solidity and hardness, seemed to
six days o f creation. Defending the account o f Moses in Genesis, Philoponus invite an explanation and thus to provide an occasion for the expression ot
insisted that the latter had given a better explanation o f the firmament than opinions about its possible hardness or softness. Few, however, chose to
either Plato or Aristotle. Whereas Plato had assumed a heaven composed avail themselves o f an opportunity to explain why, in Genesis, the term
o f the four elements and Aristotle had invoked the existence o f a fifth firmamentum was used for the heaven created on the second day.41 T w o who
corporeal element, the ether, Moses, by contrast, had assumed that the did were Vincent o f Beauvais and Campanus o f Novara. Vincent declared
firmament was formed in the midst o f the waters. Because the substance that the term was used because that heaven is ungenerated and incorruptible
o f the firmament is transparent and water and air are the only tw o elements rather than because it is immobile with respect to place. It is indeed indis­
that possess this property, Moses assumed that the transparent heavens are soluble, because it lies beyond the action and passion ot contraries.4* Cam ­
formed o f these two elements, though composed perhaps more o f water panus explains that the firmament is so called because “ its motion always
than air. The term firmamentum, implying solidity, also suggests that these seems to be firm and uniform and because the fixed stars seem to be firmed
tw o elements were transformed from their natural fluid state to a solid, and in it.” 43 B y describing the fixed stars as being “ firmed” in the firmament,
presumably hard, body.36 Campanus is perhaps implying the existence o f a hard firmament. The
Despite the seeming ambivalence or inconsistency o f Augustine and Basil, brevity o f his discussion makes judgm ent uncertain. Nowhere, however,
and Philoponus’s explicit support o f solidity, most Christian authors and do Vincent or Campanus explicitly associate solidity or hardness with the
Latin Encyclopedists during late antiquity and the early Middle Ages, and term firmamentum. Centuries later, Bartholomew Amicus also argued that
even into the thirteenth century, probably thought o f the heavens as fiery the term implied firmness and solidity.44 For Amicus, however, and all
or elemental in nature, and therefore fluid. In this category, Christopher other scholastics o f the seventeenth century, solidity signified hardness. By
Scheiner included Gregory o f Nyssa, Chalcidius, Isidore o f Seville, John of contrast, his seventeenth-century contemporary Roderigo de Arriaga in­
Damascus, Peter Damian, Hugh o f Saint Victor, Peter Lombard, Alexander sisted that nothing could properly be deduced from the term firmamentum
o f Hales, and Bonaventure,37 to which we may add Macrobius, Michael about the solidity or hardness o f the heaven called by that name.45 Just
Scot, Robertus Anglicus, and perhaps Peter o f Abano.38 A notable exception because this particular heaven is called “ firm” (firmus) does not warrant an
was Robert Grosseteste, who described the heavens not only as the most inference o f hardness, because the heavens could be “ firm, stable, and in­
subtle o f all bodies, but, relying on the words o fjo b 37.18, also characterized corruptible, even if they are not hard.” 46
it as the most solid o f bodies, like fused metal.30 Others undoubtedly offered Our information, such as it is, derives from brief statements in a variety
no opinion at all. For some o f these individuals, Scheiner inferred belief in of contexts. During the late Middle Ages, the hardness or softness o f the
a fluid heavens because they assumed that the Sun, or in certain instances celestial orbs was not judged a significant topic, as evidenced by the fact
even all the planets, were fiery bodies.40
(628, col. x); Isidore ot'Seville (634, col. 1); John Chrysostom (628, col. 1); Peter Damian
word ‘seen,’ namely, where the Scripture says: 'The birds ofthe heavens,’ and again, ‘the (629, col. 1); Hugh o f Saint Victor (631, col. 2); Peter Lombard (629, col. 2); Alexander
flying creatures below the firmament o f the heavens.’ ” o f Hales (629, col. 2); Bonaventure (629, col. 2), and others.
36. Here I follow Duhem, Le Systeme. 1913-1959, 2:499-500. 41. Although Bonaventure, for example, discussed the firmament in a few questions in his
37. Scheiner, Rosa Lrsina, bk. 4, 1630, 627—635. Scheiner devoted the fourth and dual book Sentences, bk. 2 (Opera, 2:338-341, 351-352), he nowhere considers why the term fir­
o f his Rosa Ursina to the themes o f fluidity and corruptibility o f the heavens by citing mamentum was used to describe the one or more heavens embraced by it. The same may
passages trom numerous authors who, in his judgment, had expressed explicit or implicit be said about Richard o f Middleton in the latter’s second book o f his Sentences.
opinions. Scheiner withheld his own opinion, declaring at the end o f this lengthy section, 42. “ Nos autem dicimus ad primum quod firmamentum dicitur a firmitate naturae quia non
that he would give his opinion at another time and in another place. Ibid., 773, col. i- generatur, nec corrumpitur et non ab immobilitate secundum locum. . . . Et propter hoc
38. For Michael Scot and Robertus Anglicus, see Section II o f this chapter; for Peter o f Abano, dicitur firmamentum quia indissolubilis est concensus ille cum extractus sit extra actionem
see Duhem, Le Systeme, 1913-1959, 4:253, where Duhem, in discussing Peter’s Lucidator. et passionem contrariorum. ” Vincent o f Beauvais, Speculum naturale, bk. 3, ch. 102, 1624,
says that in the latter Peter assumes that each planet moves within the medium of a fluid i:col. 230.
substance which constitutes its heaven. In her edition o f the Lucidator, Vescovini Fedenci 43. O f the firmament, which he identifies with the eighth sphere, Campanus says: “ Et dicitur
seems to arrive at the same conclusion about Peter’s acceptance o f self-moved planets, firmamentum quoniam ipsius motus semper videtur esse firmus et unitormis et quia
but she makes no claims about 1 fluid medium (Lucidator, 1988, 269—270). Indeed. Peter stellae fixe videntur firmari.” These words appear in Campanus, Sphere, 1531, I96r.
makes no mention o f a fluid medium. 44- For more on this point, see Section VIII.2 o f this chapter.
39. £st itaque celum corpus primum . . . quia subtilissimum; et tamen, ut dicit lob, solidissimum 45- “ Vides ergo ex nomine firmamenti nihil plane de soliditate deduci.” Arriaga, De caelo,
quasi ereJusutn.” Grosseteste, Hexaemeron, part. 3, ch. 16, 2, 1982, 117. For other references disp. 1, sec. 3, subsec. 4, 1632, 502, col. 2, par. 37.
to this frequently cited passage from Job, see note 51, this chapter. 46. “ Ubi caeli dicuntur firmi, nihil de soliditate contineri; sunt enim firmi, stabiles, et in-
40. In this group, Scheiner (Rosa L’rsina) includes Saint Basil (627, col. 2); Gregory ot Nyssa corruptibiles, etiamsi non duri.” Ibid., par. 38.
338 THE C E L E S T IA L REGION ARE THE H E A V E N S HARD OR FLUID? 339

that scholastics did not see fit to devote a questio to that theme. W hy did surface o f its eccentric deferent. Indeed, the planet itself is also contiguous
they ignore this question, which became so important in the seventeenth with the concave and convex surfaces ot the epicycle that surrounds it.
century? Perhaps because Augustine thought a Christian’s interpretation o f Richard was probably one o f the first in the Latin West to present the
the meaning o f “ firmament” was relatively unimportant,47 an attitude that case for heavens composed o f solid and hard orbs. Although he gives no
was reinforced by Aristotle him self who ignored the issue and provided clear indication as to which o f the two alternatives he preferred, it is likely
no guidance. that the hard-orb hypothesis was itself derived from some earlier account. 50
The hypothesis o f fluid heavens, which went largely unchallenged prior The justification from Job 37.18 was the most explicit biblical support
to the thirteenth century, came to have a rival after the introduction and available for belief in hard orbs and would be frequently cited - especially
dissemination o f Aristotelian—Ptolemaic astronomy and cosm ology in the in the seventeenth century — in defense o f hard orbs and against those who
thirteenth century. Whereas previously the idea o f fluid or soft heavens was believed in fluid heavens. ' 1 In this passage, Elihu asks Job whether, like
overwhelmingly dominant, the existence o f orbs and their possible hardness God, he could fabricate the heavens as if they were made o f molten metal, 54
now emerged as an opposition hypothesis. thus implying that God had made the heavens hard like metal.
Richard o f Middleton illustrates the change that had occurred. During Despite the general absence o f detailed and explicit discussions o f the
the course o f a discussion on whether the planets, or heavens, form one hardness or fluidity o f the celestial orbs in the Middle Ages, some scholastic
continuous body, Richard describes the tw o rival interpretations, without authors give evidence o f having assumed the hardness o f the spheres. During
choosing between them.48 The first opinion assumes that the heavens are the course o f the fourteenth century, Themon Judaeus and Henry o f Hesse,
o f a fluid nature, which Richard identifies with Aristotle’s fifth element, or and in the late fifteenth century Hartmann Schedel, explicitly argued for
ether. In this fluid theory, no distinction is made between orbs on the basis hard spheres, while in the fourteenth century Nicole Oresme did so indi­
o f different forms nor on the basis ot any discontinuity o f their surfaces, rectly, as perhaps did Pierre d’A illy in the early fifteenth century. Let us
as, for example, one stone is distinguished from another. Celestial orbs now consider the manner in which these few individuals indicated their
differ only because o f their diverse motions. But these diverse motions do preference for hard orbs.
not produce discontinuiutv in the fluid medium through which the planets In his questions on the Meteorology, Themon debated whether the sky or
can move readily and easily. Indeed, Richard may mean here that the orbs heavens are o f a fiery nature and rejected the possibility, arguing that if the
and planets are themselves fluid parts o f the overall ether and that celestial heavens had an elemental nature, they would be like earth and water rather
motions consist o f parts ot the ether m oving in different directions, as we than fire. This is because “ a heaven [i.e., orb] is a hard [durum] body without
can observe in water, “ when its different parts move to different positions.” capacity for flow ing.” 53 But “ fire in matter proper to it is not hard and
Under these circumstances, the water does not lose its form .4'4 lacking in the capacity to flow, as is obvious by experience, as we see in
Others, however, present a second opinion in which “ the fifth body flames. [Experience] also shows [that it is quite otherwise with] water, ice,
consists o f celestial solids that are not divisible. Thus [in] Job 37 it is said
that the heavens, which are most solid, are made o f metal. And the Phi­ 50. Perhaps from Grosseteste, who, in the first half o f the thirteenth century, had linked Job
losopher, in the second book o f De caelo, proves that the stars [planets] are 37.18 with a hard heavens (see Section V, this chapter).
51. Besides Richard, others who cited it were the Conimbricenses [De coeio, bk. 1, ch. 3, qu.
not moved from one part o f an orb to another part, because the orb would 1, art. 4], 1598, 70; Hurtado de Mendoza [De caelo, disp. 1, sec. s|, 1615, 366. col. 2:
be divided, which he assumes absurd. And also to many, it seems that if Amicus, De caelo, tract. 5, qu. s, art. 2, 1629, 273, col. 1 and art. 3, 278, col. 1: A versa,
the orbs were fluid and [therefore] divisible by nature, they should appear De caelo, qu. 32, sec. 6, 1627, 67; Arriaga, De caelo, disp. 1, sec. 3, subsec. 4, 1632, 302,
col. 1; Poncius. Sentences, bk. 2. dist. 14, qu. 2, 1A39, vol. 6. pt. 2:727, col. 2-728. col.
corruptible.” Because o f the contrary motions o f the orbs, those who 1; Compton-Carleton [De coeio, disp. 1, sec. 3], 1A49, 399, col. 1; Riccioli, Almagestum
adopted this interpretation denied the continuity o f the eccentric orbs and novum, pars post., bk. 9, sec. 1, ch. 7, 1A51, 240, col. 2: Bonae Spei [comment. 3, De
not only assumed their contiguity with the surfaces that surrounded them coeio, disp. 3, dub. 7], 1652, 14, col. 1; Cornaeus [De coeio, disp. 2, qu. 2,], 1657, 499;
and Oddus [De coeio, disp. 1, dub. 12], 1972. 35, col. 1 (Oddus gives the correct text,
but also assumed that the surface o f an epicycle was contiguous with the
but the wrong reference, citing Job 3.32).
52. In the Vulgate, the text reads: “ Tu forsitan cum eo fabricatus es caelos, qui solidissimi
quasi aere fusi sunt.” The Douay-Challoner translation o f the Vulgate (ed. John P. O ’Con­
47. Judging from his discussion ot the firmament in the Summit theologiae, pt. 1, qu. 68, art. nell, Chicago: Catholic Press, 1950) renders “ qui solidissimi quasi aere fusi sunt” as “ which
1, 1967. 10:71-77. Thomas Aquinas was certainly one o f these. Grosseteste thought it are most strong, as if they were o f molten brass.” A recent translation from the Hebrew
would be tedious and prolix to present a detailed analysis o f the nature o f the firmament text provides a more graphic version to describe the rigid heavens: “ Can you beat out
(see Grosseteste, Hexa'emeron, part. 3, ch. 6, 1, 1982, 106). the vault o f the / skies as he does, / hard as a mirror o f cast metal?" Sew F.nglish Bible,
48. Richard o f Middleton, Sentences, bk. 2, dist. 14, art. 3, qu. 1, 1591, 2:184, col. 1. T976.
49. It is possible that Richard is reporting a version o f the opinion of Roberrus Anglicus, as 53 - See Themon Judaeus [Meteorology, bk. 1, qu. 3 ("Utrum coelum sit nature ignis” )], 1518,
described in Section II ot this chapter. 157V, col. 2.
340 THE CELESTIAL REGION ARE THE H EAVEN S HARD OR FLUID? 341
and earth. For earth [and water] can be made hard and even transparent it follows necessarily that the concave surface o f the sovereign heaven and the convex
[perspicua], as is obvious from glass and ice.” 54*5
6Since fire cannot be hardened surface o f the second or next heaven below must be absolutely spherical, with no
it cannot be the material from which the celestial orbs are composed. roughness or humps, and that these heavens must m ove one inside the other without
Around 1390, in a commentary on Genesis (Lecturae super Genesim), Henrv anv friction. Rather, the passage o f one surface above the other must be as smooth,
o f Hesse presented an unusual interpretation o f the celestial orbs. According as gentle, and as effortless as possible. The same holds for the second and third
to Steneck (1976, 61-62), Henry argued that the firmament created on the heavens and thus through all o f them in descending order down to the concave
second day was comprised o f “ a series o f concentric shells or spheres that surface o f the lunar sphere, which is concentric with the earth and with the heavenly
stretch from the region o f the Moon to the region o f the fixed stars. Thev body which contains or comprises or is composed o f all the partial heavens; other­
are clear, firm, impenetrable, and have thickness. . . . In fact, the image of wise, all this body taken together would be thicker in one part than in another,
glass globes spinning on fixed axes around the central earth, so commonlv which is neither probable nor reasonable. Therefore the concave surtace o f the lunar
used to describe the medieval conception o f the celestial orbs, seems to fit heaven must be perfectly spherical.'7
quite nicely the discussion in the Lecturae." The various orbs had congealed
like water or lead. Henry rejected Aristotle’s celestial ether, or fifth element, The perfect sphericity o f the concave lunar surface causes the convex
and insisted that the heavenly region was composed o f matter similar to surface o f the sphere o f elemental fire to assume a perfectly spherical shape.
that o f the earth. Moreover, he further argued that on the fourth day of Ordinarily, none o f the imperfect four elements could assume a perfectly
creation the planets were formed from mixtures o f elemental matter that spherical shape. The convex surface o f fire, which Oresme describes as
rose up through the various hard celestial orbs. Because Henry believed “ perfectly polished and spherical,” is, however, an exception. But “ this is
that the movement o f such relatively coarse matter through the hard, ce­ not due to the element o f fire itself, but to the concave surface o f the lunar
lestial orbs was physically impossible, and since he was not prepared to sphere which contains the fire and which is perfectly spherical, being every­
abandon his interpretation, he chose to explain it by miraculous interven­ where in contact with the fire without intermediate plenum or vacuum .” 58
tion. Hartmann Schedel conveyed his conviction o f a hard firmament in On the basis o f Oresm e’s discussion, it seems reasonable to assume that he
his Liber chronicarum o f 1493, by a simple declaration that God “ made [the judged the celestial surfaces to be hard. Otherwise the lunar concavity could
firmament] solid, out o f water congealed like a crystal.’’5' not have shaped the outermost surface o f grosser fire into a perfectly spher­
Few were as explicit as Themon Judaeus, Henry o f Hesse, and Hartmann ical surface.
Schedel. O nly indirectly and occasionally can we inter the apparent opinions The inclusion o f Oresme among those who probably thought ot the
ot others about the hardness ot the spheres. In this connection, Nicole spheres as hard is based not on his description ot the orbs as “ perfectly
Oresme’s discussion in his French translation of, and commentary on, Ar­ polished” but on his assumption that the lunar concavity shaped the con­
istotle’s De caelo is ot interest. Here Oresm e describes the surfaces o f all tiguous, convex surface o f the perfect, fiery sphere below. Similarly, one
celestial spheres as perfectly polished and sm ooth.5'’ Because no vacua can ought not to include an anonymous fourteenth-century author among pro­
exist between any two celestial surfaces, ponents o f hard spheres simply because o f an assertion that “ celestial bodies
do not rub together in their local motions because they are highly polished.
54. Here is the full text o f Themon’s second conclusion: “ Si celum esset nature elementalis Nor is there any friction between them that could generate heat.” '9
potius esset nature aque vel terre quarn ignis. Probatur conclusio: quia celum est corpus With Pierre d’Ailly, we complete our list o f those who directly or in­
durum mfluxibile, alias fieret permixtio astrorum et stellarum nimium irregularis propter directly indicated a world o f hard celestial orbs. D ’A illy reveals his belief
divisionem eius. Sed ignis in propria sibi materia non est durus et influxibilis, ut patet
per experientiam, videmus emm de flammis. De aqua autem et glacie et terra patet. Terra in hard orbs in the context o f a conclusion in which he informs his readers
enim potest indurari etiam perspicua fieri, ut patet de vitro et glaciebus. Ergo celum that Sacrobosco had demonstrated that the intermediary spheres o f the
potius esset nature aque vel terre, quod est propositum.” Ibid.
55 - Hartmann Schedel, Liber chronicarum, pt. 1 [Rosen], 1493. Although the book is unpa-
ginated, see “ On the Work o f the Second D ay,” lines 4-5. On the opposite page, the 57. Ibid., ch. 9, 387.
Latin text reads: “ Ex aquis congelatis in modum cnstalli solidavit et in eo fixa sidera.” 58. Ibid., ch. 11, 399.
Schedel seems to have conflated the crystalline sphere with the firmament. In note 34 ot 59. “ Dicendum est quod corpora celestia in suis monbus localibus non contncantur quia sunt
this chapter, we saw that when Basil also mentioned crystalline rock and the firmament corpora polidssima. Nec inter ipsa sit aliqua contricatio talis ex qua possit gigni calor.”
in the same passage, he sought to dissociate the two. See “ Compendium ot Six Books,” Bibliotheque Nationale, tonds Latin, MS. 6752, 214V.
56. “ The primary and sovereign heaven . . . does not push nor pull the heavenly sphere which For a description o f the contents o f the treatise, see Thorndike, 1923-1958, 3:568-584.
is immediately under it. In addition, this concavity or concave area is verv completely Sometime between 1570 and 1572, Robert Bellarmine, the future Cardinal Bellarmine
and perfectly polished, planed, and smoothed so that it could not be more s o . . ■ and who played a significant role in the Galileo affair, rejected hard, polished, contiguous
without any roughness or denticulation.” Oresme, Le Livre du del. bk. 2, ch. 5, orbs by arguing that such orbs had no tendrils, braces, or glue to enable them to cling
30- to each other so that one orb could drag another with it (see Lerner, 1989, 268, n. 35).
342 THE CELESTIAL REGION ARE THE HE A V E N S H A R D OR FLUID? 343

celestial region are spherical with respect to both their concave and convex lationships o f celestial spheres, rarely did he see fit to interject an explicit
surfaces. This is true, d’A illy continues, only if we assume that “ the heavens judgm ent on their hardness or fluidity, as is evident from al- B itrujfs De
are not breakable [frangibile], [not] fluid [Jhtxibile], [not] augmentable, or motibus celorum, which Michael Scot translated from Arabic to Latin in 1217.
diminishable” and that there can be no penetration o f dimensions or exis­ Here BitrujI explains:
tence of a vacuum."0 B y denying fluidity and breakability, d ’ Ailly implies
the existence o f hard celestial orbs. It is well know n by all men that the whole heavens are composed o f mutually
different spheres and that one touches another in perfect contact. And because one
[sphere] is moved inside another, there is a hniteness o f the rotation and an equality
o f surfaces. And these [orbs] are continuous with each other because no other body
V I. O n the difficulties o f d e te rm in in g w h eth er natural
philosophers assum ed hard or fluid orbs in lies between them.
the late M id d le A ges And it is known that the concave surface o f a higher [orb] is the place oi the [orb]
next below it and between them there is neither a plenum ot another extraneous
From what has been described thus far about the problem o f the hardness
body, nor is there a vacuum, but one [orb] touches the other [orb] with its whole
or softness of celestial orbs, we perceive a gradual shift: the widespread
surface/2
assumption o f fluid heavens and orbs in the thirteenth century was yielding
to an assumption o f their hardness in the fourteenth century. Richard o f
Middleton described both theories but refused to choose between them. In this passage, BitrujI gives no indication o f his opinion on the hardness
Shortly after, Aegidius Romanus proposed a combinatory theory in which or softness o f the celestial orbs. But if the orbs are contiguous, as is likely
the overall heaven was hard but the eccentric deferents were fluid (see (see Ch. 13, Sec. III.2), or even if they are assumed continuous, does this
Ch. 13, Sec.III.9). As the fourteenth-century progressed, Them on Judaeus alone provide a clue about his opinion concerning the hardness or softness
and Henry o f Hesse explicitly opted for hard orbs, and Nicole Oresme o f the surfaces o f the celestial orbs? It does not. If an author simply opted
did so implicitly. Explicit defenders o f fluid orbs or heavens have yet to for continuity or contiguity without providing any additional clues about
turn up in the late fourteenth and fifteenth centuries. B y the late sixteenth hardness or softness, we would have no good independent reasons tor
century, the hardness theory had become explicit and widespread. Tycho assuming either hard or soft orbs. In Chapter 13 (Sec. III.2), we saw that
Brahe acknowledged it as the major opinion he had to destroy. Even as Campanus o f Novara probably assumed contiguous orbs in order to avoid
the existence of hard orbs was losing support in the seventeenth century, the possibility o f the overlap o f two successive orbs or the possibility ot
it remained the rival theory to the concept o f a fluid heaven. void space between them. Because these are impossibilities in Aristotelian
Although in retrospect the hardness or softness o f the celestial orbs appears natural philosophy, the modern editors o f Campanus’s Theorica planetarum
to us an important cosmological problem, our medieval predecessors seem infer that Campanus “ supposes the spheres to be solid,” '’3 by which, it
not to have shared that judgment. They were quite content to discuss the seems, they mean hard. If so, the inference is unwarranted Even it the
celestial orbs without any explicit, or even implicit, indication as to their spheres were composed o f a single fluid substance, the same objection would
hardness or softness. The issue rarely surfaced. This reluctance was not obtain: the overlapping parts o f the two spheres could not possibly occupy
confined to Latin scholastics but is equally apparent in authors who make the same place simultaneously. Thus Campanus’s spheres may indeed be
up the Greco-Arabic tradition. Aristotle himself never saw fit to raise the solid, but whether that solidity is to be associated with hard or sott surfaces
problem. N or did Ptolemy in the Almagest, not even in the first book, and orbs cannot be determined.
Whereas Campanus and most others provided no useful clues tor deciding
where he presents a modest amount o f information about the nature o f the
physical w orld/” Even when an author specifically discussed the interre-6 1*
0 the issue, Robertus Anglicus, who really seems to have assumed the con­
tinuity o f successive surfaces, presents explicit information that enables us
60. “ Quinta conelusio est quod ratio autoris bene demonstrat caelum esse sphaericum quan­
to conclude that he considered the orbs fluid. In order to permit the different
tum ad superficies tarn concavas quam convexas sphaerarum intermediarum. Patet con-
clusio supposito quod caelum non sit trangibile, fluxibile, augmentabile, nec diminuibile; “ assumes that the [celestial heavens] are solid and hard bodies and mutually distinct
supposito etiam quod non possic esse penetratio dimensionum nec vacuum.” D ’Aillv. 14 (videtur Ptolomaei I Almeg.c. 1, ubi supponit esse corpora solida et dura inter se distinctai.
Questions, qu. 5, 1531, 153 v. 02. M y translation from BitrujI, De motibus celorum [Carmody], 1952, S2. The words “ there
61. During the seventeenth century, it was not uncommon for scholastic authors to classify is a hniteness o f the rotation” are an uncertain rendition ot the Latin “ ideo ipse est in fine
earlier authors as proponents o f hard or fluid heavens. These identifications were some­ rotationis.” For an English translation based on Arabic and Hebrew versions ot BitrujI s
times arbitrary and without foundation, as when Amicus insisted (De caelo, tract. 5, qu. treatise, see BitrujI, On the Principles ot Astronomy, 1971, 65-66.
5, art. 2. 1626. 27s, col. 2) that in the first chapter o f book 1 o f his Almagest, Ptolemy 63. Campanus o f Novara, Theorica planetarum, 1971, 412, n. 47.
344 THE CELESTIAL REGION ARE THE H E A V E N S HA RD OR F LU ID ? 345

orbs to have independent motions and to have continuous and identical


outer surfaces, Robertus assumed that the outer surfaces o f celestial orbs V II. W hen did “ solid o rb ” b ecom e syn o n y m o u s
were immobile but that their inner parts, which he likened to a fluid, were w ith “ hard o rb ” ?
capable o f motion/’4 Because his world system required an intersection between the orbits o f
The use o f the contiguity theory o f distinct and separate celestial surfaces Mars and the Sun, which would have been impossible if hard spheres ex­
that are everywhere in perfect contact to avoid the possibility o f void isted, Tycho Brahe used his knowledge that the comet o f 1577 was moving
space or extraneous matter between them may seem at first glance to in the celestial region beyond the Moon to deny the existence o f solid
indicate that the surfaces are hard. If the convex surface o f the planetarv celestial orbs and to suggest instead that the heavenly region was composed
sphere o f Mars were contiguous to the concave surface o f Jupiter’s orb, o f a fluid substance.66 The solid celestial orbs whose existence Tycho denied,
and if those tw o distinct, touching surfaces moved with different speeds were, o f course, o f the hard and rigid variety. In 1588, he explained that
and perhaps even in different directions, one might argue that they would he “ first showed and clearly established that by the motions o f comets they
retain their separateness and move independently o f each other only be­ [the heavens] are fluid and that the celestial mechanism is not a hard and
cause their surfaces were hard. But because o f the attributes traditionally impervious body filled with various real orbs, as has been believed by many
assigned to the celestial ether, the conditions just described do not bv up to this point, but that it is very fluid and simple, with the orbits o f the
themselves warrant the inference o f hard spheres. They are equally com­ planets free, and without the efforts and revolutions o f any real spheres.” 67
patible with soft and fluid spheres. After all, not only is the celestial ether From T y ch o ’s assertions that “ many” or “ very many” contemporaries be­
unalterable and incorruptible, but it was usually assumed to be more lieved that the heavens were composed o f hard, celestial orbs, we may infer
subtle than air and fire.665 With such properties, the ethereal orbs could
4 that this was the comm only held opinion o f his day. T ych o ’s influence was
be contiguous but fluid. Moreover, because o f their presumed unalterable so great that seventeenth-century astronomers and natural philosophers who
and incorruptible natures, fluid orbs could be in contact at every point mentioned solid spheres, whether or not they agreed with him, did so with
and also retain their perfect spherical shapes. Perfect, incorruptible, con­ the understanding that they were hard and rigid.
tiguous fluid surfaces, no less than hard surfaces, could be polished and Alm ost from the first formulation o f Tycho's radical interpretation, scho­
move without resistance or friction. If I earlier attributed to Oresme an lastic natural philosophers found themselves divided. As a direct reflection
implicit belief in hard spheres, it was not because o f his assumption o f o f that division o f opinion, scholastic authors introduced a new question
polished surfaces moving without resistance, but rather because o f his into their commentaries on Aristotle’s De caelo, one that was unknown to
declaration that the lunar concavity shaped the contiguous, convex surface the Middle Ages. They asked whether the heavens are solid or fluid. Al­
o f the fiery sphere below it. though some sided with Tycho and assumed fluid heavens without spheres,
Because the celestial ether was traditionally assigned properties that whereas others defended the existence o f hard spheres and still others argued
made it appear rare and fluid-like, it might seem that, in the absence o f for a combination o f hard and soft, almost all were agreed that a solid sphere
an explicit assertion in favor o f hard spheres, or indirect independent signified a hard sphere.68 Giovanni Baptista Riccioli underscores the pow-
evidence indicating rigid spheres, it would be more plausible to assume
that a medieval author considered the celestial orbs solid and soft rather
66. On this, see Thoren, 1979, 53—<>7.
than solid and hard. Such a judgm ent would also be unwarranted and 67. “ Ubi per Cometarum motus prius ostensum et liquido comprobatum tuerit. ipsam Coeli
misleading. In the absence o f reasonably compelling evidence, it is wiser machinam non esse durum et impervium corpus varijs orbibus realibus confertum. ut
to draw no inferences. hactenus a plensque creditum est, sed liquidissimum et simplicissimum. circuitibusque
Planetarum liberis, et absque ullarum realium Sphaerarum opera aut circumvectione."
Tycho Brahe, De tnundi aetherei, 1922, 4:59. On page 222 o f the same work, Tycho savs
much the same thing, emphasizing that “ very many modern philosophers . . . distin­
64. See Chapter 13, Section III.2 and note 67. guished the heavens into various orbs made of hard and impervious matter” (et recentiores
65. Aristotle nowhere says this, but it seems to follow from his ordering o f the four elements etiam Philosophos quamplunmos, qui Coelum ex dura et impervia materia Orbibus varijs
(earth, water, air, and tire), the latter three ot which become rarer and more subtle as distinctum); for a similar statement, see also page 223. T ycho’s De rnundi aetherei was
their distance from the earth increases. Since the celestial ether extends beyond tire, it reprinted in 1603 and 1610 and was thus widely known. For these and other references
should exceed the latter in rarity and subtlety. In the strict sense, o f course, Aristotle to Tycho, I am indebted to my late colleague. Professor Victor Thoren.
denied that contrary qualities such as rarity and density could be applied to the ether. But 68. For example, see Arriaga, De caelo, disp. 1. sec. 3, 1632, 499. col. 1—504, col. 1 (“ Whether
because physical attributions were sometimes unavoidable, the celestial ether was likely the heavens are incorruptible and solid” ); Serbellonus [De caelo, disp. 1, qu. 2, art. 4],
to be considered purer, rarer, and more subtle than any other substance. Even planets, 1663, 2:25, col. 1-28, col. 1 (“ Whether the celestial bodies are solid or liquid” ); Thomas
which were usually conceived as aggregations o f ether sufficiently dense to reflect light Compton-Carleton, De coelo, disp. 1, sec. 3, 1649, 398, col. 2-399, col. 2 (“ Whether the
and become visible to us, were hardly thought o f as “ dense” in the sense in which that heavens are solid or fluid” ); and Cornaeus, De caelo, disp. 2, qu. 2, dub. 3, 1657, 494-
term might be applied to any o f the four elements or bodies compounded o f them. 500 (“ Whether the heavens are hard and solid").
346 THE C E L E S T IA L REGION ARE THE H E A V E N S H A R D OR FLUID? 347

erful association o f solidity with hardness when he explains that although Reisch says that it may also be called “ crystalline because it is a nonconcave,
the term soliditas means three-dimensional, it also “ has associated with it transparent, and lucid solid body [corpus solidum] in which the subtle bodies
hardness [as] opposed to softness, as we say that marble is solid, and metal, o f the blessed move without resistance or penetration of dimensions. ” ~!
as long as it does not liquefy, and even ice before it melts. " (H > Here was a Despite an attribution o f transparency and lucidity to this crystalline solid
significant departure from the Middle Ages, when, as we saw, the term body, Reisch neglects to inform his readers whether it is hard or soft.
solidum could signify either hardness or softness. The powerful bond that was forged between hardness and solidity in the
But when did a solid orb come to imply a hard and rigid orb? From what seventeenth century was apparently not achieved by the time o f Copernicus.
has been said already, the indissoluble nexus between solid and hard prob­ Because Tycho claimed that many o f his contemporaries believed in hard
ably did not occur during the Middle Ages. Although Themon Judaeus, spheres, it is not unreasonable to assume that the firm connection between
Henry of Hesse, and Hartmann Schedel made explicit commitments to hard solidity and hardness became explicit and commonplace during the period
orbs, and a few others implied the hardness o f the celestial orbs, most offered between the emergence o f Copernicus’s De revolntionibus and T ych o ’s pub­
no opinion. Natural philosophers who commented on Aristotle’s De caelo lication o f his cometary researches in the late 1580s. H ow this occurred and
and theologians who commented on Peter Lombard’s Sentences either did who might have been instrumental in its development are unknown, and
not think about the problem at all or, if they did, felt no compulsion to I shall not pursue this further except to suggest that Clavius does not seem
discuss it. Astronomers were no different. Thus ibn al-Haytham (Alhazen) to have been the disseminator o f hard and solid celestial spheres, as has been
and Georg Peurbach, whose treatises played instrumental roles in dissem­ argued.7+
inating knowledge about solid spheres during the late Middle Ages and the Indeed, despite T ych o’s assertion about his contemporaries, another hy­
Renaissance, undoubtedly spoke about real, material orbs, but neither gives pothesis may be equally plausible; that the firm connection occurred after
any indication whether those orbs are rigid or fluid.6 70*7
9 2Although it is likely T y ch o ’s publication o f 1588 and because o f it, not between the time o f
that Copernicus believed in the existence o f solid spheres, it does not follow Copernicus and Tycho. B y showing that the paths o f comets and the in­
that he, “ like every other astronomer o f his time, envisioned planetary tersection o f the orbits o f the Sun and Mars made hard orbs an untenable
models to be composed o f non-intersecting, rigid spheres. Nothing that assumption and virtually compelled acceptance o f an orbless, fluid heavens,
Copernicus said or implied in the De revolntionibus enables us to decide with Tycho himself forced the issue. To Tycho and his followers, the denial of
any confidence whether he assumed hard or fluid spheres. Copernicus fits fluiditv left only one meaningful option: hard orbs. In the debate that fol­
the pattern of the Middle Ages, when explicit opinions about the rigidity lowed, the word “ solid” came somehow to be inextricably linked with
or fluidity of the orbs were rarely presented. hardness, which, in turn, came to be the major qualifier o f the term “ orb.”
Approximately one-half century before Copernicus published the De re- Whereas once the word “ solid” signified any fluid or hard body that was
volutionibus, the same pattern is revealed in a widely read sixteenth-century o f a continuous nature - that is, without vacua - later the idea o f solid came
encyclopedia. In the Margarita philosophica, Gregor Reisch describes a sphere to be opposed to permeability and penetrability and was instead linked to
as “ round and solid’’ and characterizes it as “ a solid body contained bv a hardness. The divorce o f solidity from fluidity and its nexus with hardness
single surface.” 7" In a rather strange depiction o f the empvrean sphere, was not a necessary development but may have occurred as a consequence
o f T v ch o ’s contributions and influence. But the connection may not have
69. Controversia igitur est de soliditate presse sumpta quae praeter tnnam dimensionem taken hold until the seventeenth century. In any event, Clavius seems to
habct adiunrtam durniem mollmei oppositam. quomodo soiida dieimus esse marmora et
have ignored it in the numerous late sixteenth- and early seventeenth-
metalla quamdiu non liqueseunt et ipsa glaaes antequam dissolvatur." Riccioli, Alma-
gestum novum, bk. 9, sec. i. ch. 7, iftsi. 23s, col. a. century editions o f his commentary on Sacrobosco’s treatise. In these edi-
70. Svverdlow, 1976, who assumes that solid orbs or spheres were rigid, implies that in both
the Latin translation of ibn al-Haythams ‘On the Shape o f the Universe," which Swer- Vel sphera est corpus solidum unica superflcie contentum.” Reisch, Margarita philosophica.
dlow claims introduced “ solid sphere planetary models into medieval European astron­ bk. 7, ch. 1, 1517, 242; see pp. vi-vii For a brief summary o f Reisch’s life.
omy (117), and in Peurbach s Theoricae novae planetarum. the solid spheres were assumed 73. “ Dicitur etiam . . . crystallinum eo quod sit corpus solidum non concavum transparens et
hard and rigid (see 109—110, 116—117). I have examined both treatises and encountered lucidum in quo tamen corpora beatorum subtilia sine resistentia aut dimensionum pen-
only the usual silence on the issue o f hardness or softness. For ibn al-Havtham’s treatise, etratione ambulabunt.” Ibid., ch. 8. 247.
see ibn al-Haytham, 194a, 285-312; for Peurbach. I have used the edition included in 74. Donahue, 1975, 263. Although Clavius assumed material and physical orbs and even says
d Ailiy s Spherae traaatus (Venice, 1531)- As we saw earlier. Camp auis o f Novara also that “ the world is indeed a solid sphere” (Mundus siquidem est sphaera soiida; Clavius
ignored the issue. [Sphere, ch. 1). Opera. 1611. 3:9). to my knowledge nowhere in his Commentary on the
~i. Swerdlow, 1976, ion- 109. In the seventeenth century, Cornaeus (De caelo, disp. 2, qu. Sphere of Sacrobosco does he explicitlv characterize them as hard nor even imply it. Indeed,
2, dub. 3, 1657, 499) included Copernicus among those who believed in fluid heavens. like most o f his predecessors, he does not raise the issue. By “ solid sphere ’ Clavius meant
72. "Sphera est tale rotundum et solidum quod ab arcu semicirculi circumducto describitur. only that the world is a plenum.
348 THE CELESTIAL REGION ARE THE H E A V E N S H A RD OR FLUID? 349

tions, he proceeded like most o f his medieval predecessors: he defended the Francesco Patrizi, who, as early as 1591, insisted that the heavens were
existence o f solid orbs but found no reason to indicate whether they were neither hard nor solid. H owever, it seems unlikely that Patrizi was influ­
hard or soft. enced by Tycho, since he relied not on astronomical arguments but on
Whether orbs were hard or fluid, however, was much less important than metaphysical and general cosmological claims. For Patrizi, the basic building
whether planet-bearing orbs existed at all as opposed to a celestial region blocks o f our universe are things like light, heat, and space, none ot which
devoid o f orbs but filled entirely with fluid substances. The new celestial are hard or solid.77 But if nothing in the heavens is hard or solid, then the
discoveries - especially comets and planetary satellites - were not only heavens could not support stars or planets that are fixed in it like knots.78
destined to affect which alternative was chosen but even prompted solutions Patrizi concludes that “ all Philosophers and Astronomers err who teach that
that incorporated both. the planets [stellae] are fixed in the heavens like knots in tables.” 79
Belief in the existence o f solid spheres was nevertheless common in the
late sixteenth and the early seventeenth century. B y 1630, however, if not
earlier, it seems to have lost its dominance, lingering on as a minority
V III. T h e scholastic reactio n to T y c h o B rahe: hard orbs or
fluid heavens, or b o th , in the late sixteen th and opinion.80*In his Cursusphilosophicus o f 1632, Roderigo de Arriaga explained
the seven teenth century? that just a few years earlier celestial incorruptibility and hard solidity “ were
absolutely beyond controversy.” 8' B y the time his book appeared, fluid and
In Chapter io, we considered at some length the nature o f celestial incor­ corruptible heavens had largely replaced the two previously entrenched
ruptibility. Tycho Brahe’s investigations o f the new star o f 1572 and the concepts and had done so because “ o f the diligent observations o f certain
comet ot 1577 directly challenged that long-held, powerful Aristotelian mathematicians and astronomers, which [observations] were discovered
concept. N ot only did Tycho repudiate solid and hard orbs, but he replaced with the aid o f new and excellent instruments, especially the telescope. Thus
them with a fluid material through which the planets moved. Dramatic did some [individuals] begin to w holly invert the structure o f the heavens. ” 8‘
celestial changes o f the kind represented by new stars and comets convinced In one important sense Arriaga was typical o f seventeenth-century scholastic
Tycho and many others that not only were the heavens composed o f a authors. Most were well aware o f the arguments by Tycho, Galileo, and
fluid, rather than hard, substance but that changes did indeed occur in the Kepler on the nature o f the heavens. As the century moved on, a scholastic
heavens, which could no longer be conceived as incorruptible. literature developed which incorporated the arguments o f these great figures
Although it was Tycho who first gave scientific arguments for rejecting o f the new science and cosm ology. Scholastic authors could thus learn
hard orbs in favor o f fluid heavens, the fluid-heavens hypothesis had been indirectly o f these arguments from members o f their own group or directly
popular in the Middle Ages prior to the introduction o f the Aristotelian from the works o f these three astronomers. In one or both o f these ways,
ether and the gradual emergence o f hard orbs. It reemerged in the sixteenth most became aware o f the crucial observations and arguments. As they did,
century at the very time when the hard-orb theory was at its height. Between they gradually abandoned hard orbs for fluid heavens, so that by 1672,
the years 1570 and 1572, Robert Bellarmine (1542-1621) emphatically re­
jected hard orbs - indeed orbs o f any kind75 - and insisted that celestial 77. “ Si vero e?t nostris fundamentis, e spado, ex lumine, ex calore, ex fluore, camquam ex
communibus rerum omnium elementis, celum dixerimus esse compositum verum equi-
bodies moved freely through a fluid medium “ like birds in the air and fish dem dicemus. sed non tanta posse esse duritie ac soliditate ut tarn vehementi rotatu. non
in the sea.” 76 dispergatur. Si vero coelum. quod Chaldaei docuerunt solum esc lumen, lumim nulla est
Another early opponent o f hard orbs and advocate o f fluid heavens was soliditas, nulla durities.” Patrizi, Painosmia, 1591, 89, col. 2.
78. “ Si nihil in coelo est durum, si nulla est in eo soliditas, nullam profecto tixionem. vel
nodorum, vel stellarum potuit suscipere.” Ibid.
75. In his Louvain Lectures o f between 1570 and 1572 (unpublished until 1984), Bellarmine 79. “ Toto ergo errarunt coelo et Philosoplu et Astronomi omnes, qui Stellas coelo hxas, uti
declared that “ such complex and extraordinary structures as epicycles and eccentrics are nodos tabulis esse docuerunt.” Ibid.
dreamed up so that that even the astrologers are reticient to speak about them.” Bellar­ 80. Donahue, 1975, 273, declares that “ by the end o f the 1620s the debate over the fluidity
mine, 1984, 22 (English), 23 (Latin). Although Bellarmine used the term astrologi, he o f the heavens was very nearly concluded.” The estimate seems reasonable. Thoren
clearly means astronomers. declares (1990. 254) that “ in the second half o f the sixteenth century at least, intellectuals
76. Bellarmine explains: “ Si assere velimus coelum sydereum non esse nisi unum, et illud in general and Tycho Brahe in particular believed that something real existed in the heavens
igneum, vel aereum: quod saepius conformius scripturis esse diximus: necessario iam to carrv the planets through their appointed rounds.” By 1587. however, Tycho came
diccre debemus. Stellas non moveri ad motum coeli, sed motu proprio sicut aves per to reject the existence o f hard orbs. Thoren. ibid., 258.
aerem, et pisces per aquam.” Ibid., 19 and 38, n. 88: also quoted in Baldini, 1984, 301- 81. “ Utrumque ante aliquot annos omnino extra controversiam tuerat. ’ Arriaga. De eaelo.
See also Lemer, 1989, 268. The brief translation is mine. Defenders o f orbs, whether hard disp. 1, sec. 3, 1632, 499, col. 1.
or soit, tound it difficult to believe that celestial bodies could be self-moved “ like birds 82. “ Propter quorumdam mathematicorum et astronomorum diligentes observationes quas.
in the air and fish in the sea.” See Chapter 13, note 16, where other users o f one or both novis exquisitisque instrumentis adiuti, invenerunt. et praecipue tubi optici subsidio.
ot these metaphors are also mentioned. caelorum structura penitus a nonnullis inverti coepit." Ibid.
350 THE CELESTIAL REGION ARE THE H E A V E N S H A R D OR FLUID? 351

when George de Rhodes published his discussion, he could say o f the pla­ priately listed with the fluid theorists, because the assumption o f fluidity in
netary heaven, “ no one now denies the fluidity o f the heaven o f the the heavens, whether for all or part o f it, marks a strong departure from
planets. ” 83 what was taken as the major opposition theory in the late sixteenth century,
Because hard orbs had been regularly linked with incorruptibility, it namely heavens conceived as totally solid and hard.
seemed natural to associate fluid heavens with corruptibility. Some scho­ What did terms like “ fluid” and “ liquid” mean to opponents o f hard orbs?
lastics, however, found these rigid pairings unwarranted, perhaps because In responding to the question “ Whether the heavens are fluid,” Roderigo
some wished to assert fluidity and nonetheless retain the concept o f celestial de Arriaga explained that the fluid he had in mind need not be a “ watery
incorruptibility. “ Crystal, stone, w ood, etc. are solid bodies,” observes liquid” (liquor aqueus), for “ it suffices if they [the heavens] are easily perme­
Arriaga, “ but are not incorruptible, and some substance might be easily able, much like our air, which is, nevertheless, not called absolutely fluid.” 88
permeable and yet not be corrupted.” 84 He was further convinced that Thus the heavens could range from a liquid to a gas and still be categorized
“ some experiences can be adduced for proving the fluidity o f the heavens as a fluid. The meaning o f fluidity was apparently extended in this manner
which do not thereby prove their corruptibility. Contrarily, other experi­ to avoid the charge that watery, liquid heavens would fall down upon us
ences can be adduced to prove that the heavens are corruptible which cannot in the form o f rain. A vaporized fluid, analogous to air, was more readily
show that they are fluid. Thus it is necessary to distinguish between them.” 85 conceived to remain in its celestial location high above us.89
Among late sixteenth- and seventeenth-century scholastic natural phi­
losophers whose works play a significant role in this study, opinion was
divided on the issue o f fluidity or solidity. Those who defended the existence I. Scholastic arguments for fluid heavens
o f solid orbs were the Coim bra Jesuits, Bartholomew Amicus, and Thomas
Arguments for or against fluid heavens came from a variety o f sources.
Compton-Carleton, while the defenders o f fluid heavens included Pedro
Some were drawn from Scripture and the Church Fathers and were largely
Hurtado de Mendoza, Roderigo de Arriaga, Francisco de Oviedo, Giovanni
appeals to authority; others were derived from Aristotelian physics and
Baptista Riccioli, Franciscus Bonae Spei, Melchior Cornaeus, Sigismundus
cosm ology and the scholastic additions thereto. But the most significant
Serbellonus, and George de Rhodes. O f the eight defenders o f fluidity,
and most challenging drew on, or were responses to, the new discoveries,
seven published after 1632, thus strengthening the view that by 1630 most
or the “ new phenomena” (nova phaenotnena) as they were sometimes called,90*
scholastics had abandoned hard and solid orbs in favor o f fluid heavens.868 7
associated most prominently with the names o f Brahe and Galileo. We shall
O f the authors listed as supporters o f fluidity, some held a third opinion,
have occasion to consider all o f these types.
which envisioned heavens that were partly hard and partly fluid. In this
group we may place Hurtado de Mendoza, Aversa, Riccioli, and Serbel­
lonus.s_ They, and others who shared this interpretation, may be appro­
a. Scripture
83. “ Prima ergo pars de fluiditate coeli planetarum a nemine nunc negatur.” De Rhodes [De Just as the passage from Job 37.18 served as the most important biblical
coeio. bk. 2. disp. 2. qu. 1, sec. 2, pt. 2], 1671, 280, col. 1.
84. The full statement reads: “ Primo suppono non esse idem corpus esse liquidum et esse support for hard orbs, so did Isaiah 51.6 serve to uphold fluid heavens with
corruptible; neque e contrario idem esse corpus soli dum et incorruptibile: nam crystallus, the words “ quia caeli sicut fumus liquescent” (because the heavens appear
lapis, lignum, etc., sunt corpora solida et non sunt incorruptibilia; et potest esse aliqua as smoke).9' Indeed, a number o f scriptural quotations were arrayed on
substantia facile permeabilis, licet non possit corrumpi.” Arriaga, De caelo, disp. sec.
3, subsec. 1. 1632. 499, col. 1. each side o f the controversy and largely offset each other. One and the same
85. “ Hinc suppono secundo aliquas expenentias adduci ad probandum caelos esse fluidos quae
non propterea quidquam probant de eius corruptibilitate; alias vero e contrario ad pro­
bandum eos esse corruptibiles quae non ostendunt illos esse fluidos, unde eas oportet esse, planetarum autem fluidum). Riccioli, Almagestutn novum, pars post., bk. 9, sec. 1,
valde inter se distinguere.” Ibid. Amicus, De caelo, tract. 5, qu. 5, art. 1, 1626, 272, col. ch. 7, 1651, 244, col. 1.
1 also agreed with Arriaga’s position. 88. “ Tertio suppono, cum quaerimus an caeli sint fluidi non quaeri a nobis an sint quasi
86. The seventeenth century scholastic authors were not chosen for this study by virtue ot quidam liquor aqueus, qui facile labitur; sufficit emm si sint facile permeabiles ad modum
the opinions they held but largely because they seemed to include relevant discussions quo est noster aer, qui tarnen non vocatur absolute liquor.” Arriaga, De caelo, disp. 1,
and were reasonably well distributed through the century. sec. 3, subsec. 1, 1632, 499, col. 1.
87. Riccioli, for example, assumed it more probable that the heaven o f the fixed stars was 89. Illuminatus Oddus argues this way (De coeio, disp. 1. dub. 12. 1672, 35, col. 1).
solid and the heaven o f the planets fluid. Thus we read, in the enunciation o f his final 90. The expression nova phaenotnena probably meant "new celestial appearances.” For its use,
and sole conclusion, “ it is more probable, although hardly evident mathematically or see Amicus, De caelo, tract. 5, qu. 5, art. 2. [626. 273, col. 1 and de Rhodes. De coeio.
phvsicallv, that the heaven of the fixed stars is solid and that o f the planets fluid” (Prob- bk. 2, disp. 2, qu. 1, sec. 2, pt. 2, 1671, 2S0, col. 1.
abilius multo est, licet nondum mathematice aut phvsice evidens, caelum fixarum solidum 91. The Latin is from the Vulgate.
352 THE CELESTIAL REGIO N ARE THE HE A V E N S H A R D OR FLUID? 353

author might even present scriptural passages in support o f each side 929 3 passages could be cited for each side o f the controversy, thus effectively
Nevertheless, such passages were invoked because scriptural authority was eliminating them as a critical factor in the ultimate outcome, individual
still deemed important. Moreover, because each side could muster biblical authors could still be powerfully persuaded by their personal interpretations
support, it was not unusual for scholastic authors to show that the biblical o f relevant scriptural texts, as we shall see when we examine Am icus’s
passages cited by their opponents were irrelevant or inappropriate. Thus defense o f hard heavens.
although Amicus ([De caelo, tract. 5, qu. 5, art. 3], 1626, 278, col. 1) cites
the passage from Job in his argument in favor o f solid orbs, which he
believed the true opinion, he also cites an argument, in the section where b. The new discoveries
he presents the case for fluid orbs, to show that the same passage was The most significant arguments in favor o f fluid heavens were based upon
irrelevant to the case for solid, hard spheres.1,3 In this argument, we are told the observational achievements o f Tycho Brahe and Galileo Galilei, the
that the Job passage does not really attribute hardness to the heavens and former relying on the naked eye, the latter on the recently invented tele­
that the words are not those o f God or Job but o f Elihu, whose utterances scope. The cumulative impact o f their obervations and the inferences drawn
are not accepted as true.94 Indeed, Melchior Cornaeus rejected the relevance from them transformed cosmology. Comets that Aristotle had characterized
o f the passage because the words were those o f Elihu, an unlearned man, as sublunar phenomena were now placed by Tycho in the vicinity o f Mars
whom God subsequently denounces for uttering ignorant opinions (in Job in the celestial region. Tycho had also identified the new star o f 1572 as a
38.2).95 George de Rhodes went much further and simply denied the rel­ genuinely novel celestial phenomenon and thus challenged the traditional
evance o f the argument from Job, as well as three other biblical passages. opinion o f celestial incorruptibility. With his telescope, Galileo added to
Scriptural texts do not signify that the firmament is a hard, solid body, de these the satellites o f Jupiter97 and a picture o f celestial bodies that had
Rhodes insists, but only that it is “ a body that has a constant and perpetual irregularities, especially the Moon, whose mountains and valleys made it
state” ([De caelo, bk. 2, disp. 2, qu. 1, sec. 2], 1672, 280, col. 2). De Rhodes akin to the earth.98 Galileo proclaimed “ the earth very noble and admirable
even denied the relevance for either side o f Isaiah 51.6. “ When it is said,” precisely because o f the diverse alterations, changes, generations, etc. that
he argued, “ that the heavens [literally] pass away [liquescent]," this only occur in it incessantly, . . . and I say the same o f the Moon, o f Jupiter, and
signifies that they will be changed into a better state.96 o f all other world globes” (Two Chief World Systems [Drake], 1962, 58—59)-
Because they could be assembled for either side, biblical passages could Also significant for the debate about hard or fluid orbs were Galileo’s dis­
not play a crucial role in the debate over hardness and softness, as they did coveries o f the phases o f Venus and sunspots.99
in the Copernican controversy, where all relevant scriptural passages upheld
one side o f the dispute, namely the traditional interpretation o f the Sun i. Comets. O f the new discoveries, Tycho Brahe’s determination o f the
revolving around a stationary earth. Moreover, the Church never intervened celestial nature o f comets was perhaps the most dramatic event in turning
in the issue o f the hardness or fluidity o f the heavens. Although scriptural scholastic opinion from hard orbs to fluid heavens. T o appreciate the mo­
mentous challenge that T ych o ’s achievements posed to Aristotelian com-
92. For Riccioli's scriptural citations in behalf o f hardness, see his Almagestum novum, pars
post., bk. 9, sec. 1. ch. 7, 1651, 240. col. 2-242, col. r; for his citations in favor o f fluidity,
etarv theory, and therefore to Aristotelian cosm ology, it is necessary to
see ibid., 242, col. 1—244, col. 1; lor Amicus s citations ot biblical passages in favor ot describe briefly Aristotle’s theory o f comet formation as expressed in his
fluidity, see his De caelo. tract. 5, qu. 5, art. 2, 1626, 272, col. 2-275, col. 2 and in favor Meteorology. At the beginning o f the latter treatise, Aristotle declares that
ot solid hardness see 275, col. 2-278, col. 1. Arriaga cites the Job and Isaiah passages in
De caelo, disp. 1, sec. 3, subsec. 4, 1632, 503, col. 1, par. 39. the region o f the world with which meteorology is concerned is “ nearest
93. However, this was not his real opinion, as will be seen shortly. to the motion o f the stars” ( 1 .1.338b.20-22 [Webster], 1984), by which he
94. Amicus, De caelo, tract. 5, qu. 5, art. 2, 273, col. 1. De Rhodes, De coelo. 1672, bk. 2, meant the region o f air and fire just below the M oon. The kinds o f phe­
disp. 2, qu. 1, sec. 2, pt. 2, 280, col. 1, argues that these are only the words o f Job's
friend and also asserts that the words about the metallic solidity o f the heavens are meant
nomena that occur in this region include comets, meteors, and the M ilky
to apply to an immense extent o f air. Way. Comets are thus not celestial phenomena but occur in the upper
95. Cornaeus, De coelo, disp. 2, qu. 2, dub. 3, 1657, 499. Franciscus Bonae Spei, De coelo,
comment. 3, disp. 3, dub. 7, 1652. 14, col. 1, had, a few years earlier, used the same 97. Galileo also mistakenly identified the rings o f Saturn as satellites. See Drake’s article in
argument and the same appeal to Job 38.2. Without specifically citing Job 38.2, Serbel- Dictionary of Scientific Biography, 1970-1980, 5:241, col. 2. As a consequence, scholastics
lonus, De caelo, disp. 1, qu. 2, art. 4, 1663, 2:25, col. 2, repeated the same argument as occasionally mentioned Saturn's “ satellites.”
Cornaeus and Bonae Spei. 98. Galileo described the lunar irregularities and the satellites ofjupiter in The Starry Messenger
96. “ Cum dicitur quod coeli liquescent, sigmfleatur tantum quod mutabuntur in meliorem (Sidereus nuncius) o f 1610. For translations, see Galileo [Drake], 1957, 21-58, and Galileo
statum.” De Rhodes, De coelo, bk. 2, disp. 2, qu. 1, sec. 2, pt. 2. 1671, 280, col. 2. The [Van Heldenj, 1989.
relevance o f this passage rested wholly on the term “ liquescent," which, in the context, 99. See Galileo’s Letters on Sunspots (1613) in Galileo [Drake], 1957. 87-144. Galileo discovered
did not even mean “ liquify" but rather signified “ pass away” or “ melt away.” the phases o f Venus in 1610. after he had written the Starry Messenger.

i
354 THE C E L EST IA L REGION ARE THE H E A V E N S H A R D OR FLUID? 355
atmosphere between earth and Moon. In the fourth chapter, Aristotle savs from below and meets it. The kind of comet varies according to the shape which
(ibid., i . 4.341b.7—25) the upper region is comprised o f two kinds o f ex­ the exhalation happens to take.'00
halations:
During the Middle Ages, most theories about comet formation remained
close to Aristotle’s account. If they diverged, it was not with respect to the
One kind is rather ot the nature o f vapour, the other o f the nature o f a windv
sublunar location o f comets. By placing the comet ot 1577 in the celestial
exhalation. That which rises from the moisture contained in the earth and on its
region and doing so on the basis o f carelul observation o f the comet’s
surface is vapour, while that rising from the earth itself, which is dry, is like smoke.
parallax, Tycho changed forever the debate about comets, as is readily
O t these, the windv exhalation, being warm, rises above the moister vapour, which
apparent by a glance at Riccioli’s lengthy section on comets in his Almagestum
is heavy and sinks below the other. Hence the world surrounding the earth is ordered
novum o f 1651,,0JI where he summarized virtually all the relevant arguments
as follows. First below the circular motion comes the warm and dry-element, which
with respect to the formation, substance, location, and distance o f com ets.102
we call fire, for there is no word fully adequate to every state o f the sm oky evap­
Opinions on the location o f comets ranged trom below the Moon, to above
oration; but we must use this term inology since this element is the most inflammable
the Moon, and to some comets below and some above. Theories about the
o f all bodies. Below this comes air. We must think o f what we just called fire as
matter from which comets were formed varied from the sublunar elements
being spread round the terrestrial sphere on the outside like a kind o f fuel, so that
in various manifestations to celestial matter either by means o f condensation,
a little motion often makes it burst into flame just as smoke does; for flame is the
by the alteration o f parts o f the heavens, and even by matter flowing from
ebullition o f a dry exhalation. So whenever the circular motion stirs this stuff up
the Sun and planets themselves.103 Toward the end o f what was surely one
in any way, it catches fire at the point at which it is most inflammable. The result
o f the lengthiest and most detailed studies o f comets in the seventeenth
differs according to the disposition and quantity o f fuel.
century, Riccioli arrived at certain cautious conclusions that conceded only
the probability, but not the certainty, o f supralunar comets. Because he was
From this physical arrangement o f the upper atmosphere, Aristotle explains not yet convinced that there had been any absolute demonstration that any
(ibid., 1.4.342a. 16-30) the formation o f various meteoric occurrences, in­ comets had occurred above the M oon ,104 Riccioli concluded in favor o f the
cluding comets: probability that some comets occurred above the Moon and some below.
History, he acknowledged, could furnish no information to help determine
When the phenomenon is formed in the upper region it is due to the combustion
cometarv locations.!0:i It was not only that comets moved across the heavens
o f the exhalation. When it takes place at a low er level it is due to the ejection o f in ways that made the existence o f hard orbs difficult to defend - comets
the exhalation by the condensing and cooling o f the moister exhalation; for this were thus frequently invoked in support ot fluid heavens, as Melchior C or-
latter as it condenses and inclines dow nw ard contracts, and thrusts out the hot
naeus argued100 - but the typical theory o f comet formation also made solid,
element and causes it to be thrown downwards. . . . So the material cause o f all these hard orbs seem impossible. Thus Sigismundus Serbellonus, who agreed
phenomena is the exhalation, the efficient cause sometimes the upper motion, some­
with Aristotle that sublunar exhalations could produce comets,107 was con-
times the contraction and condensation o f the air. Further all these things happen 100. See also Jervis. 1985, 11-13.
below the moon. 101. Riccioli devoted the eighth book to comets and new stars.
102. In Almagestum novum, pars post., bk. 8, sec. 1. ch. 23. 1651, 117, col. 2—120. col. 1.
Riccioli cites the opinions o f others on the place, parallax, and distances o f comets from
Aristotle assumed that the dry and warm, or fiery, exhalation and a great the earth.
103. Ibid., ch. 13, 57. col. 2-58. col. 2.
part o f the air below it is carried circularly around the earth by virtue o f 104. Under the heading “ Conclusiones de distantia et loco cometarum," the second conclusion
the circular celestial revolution. In the process o f being carried around, and reads: “ Nullus adhuc cometarum demonstratus est absolute fuisse supra lunam, sed ex
under the right conditions, parts o f the dry and warm exhalation or the hvpothesi tantum probabili quidem, sed tamen incerta.” Ibid., ch. 23, 119, col. i.
105. Fourth conclusion: "Probabile est aliquos cometas fuisse supra lunam, aliquos vero infra,
upper air might ignite. “ We may say, then,” Aristotle continues (ibid. etiam ex illis de quorum loco ex nuda historia nihil constat." Ibid., col. 2.
1.7.3443.15-21), 106. He asserts that "the comet, which we saw in 161S, was, according to the common
opinion o f astronomers, in the heaven itself. Theretore the heaven is not hard, but
permeable and fluid, like air" (Cometes ille quern anno ibi8 vidimus communi astron-
that a comet is formed when the upper motion introduces into a condensation ot omorum consensu intra ipsum coelum fuit. Ergo coelum non est durum, sed permeabile
et liquidum ad instar aeris). Cornaeus, De eoelo. disp. 2. qu. 2, dub. 3, 1657, 499.
this kind a fiery principle not o f such excessive strength as to burn up much o f the
107. In opposition to Aristotle, however. Serbellonus also believed that comets could be
material quickly, nor so weak as soon to be extinguished, but stronger and capable celestial and that thev could be produced by effluences given off by planets. In short,
ot burning up much material, and when exhalation o f the right consistency rises they could also be produced by celestial matter.
356 THE CELEST IAL REGION ARE THE HE A V E N S H A R D OR FLUID? 357

vinced that comets demonstrate fluid heavens. N ot only are comets visible ii. New stars. We saw earlier how scholastics o f the sixteenth and seven­
below the Moon, but they are also seen above the Sun, Mars, and Saturn teenth centuries coped with the problem o f new stars in their efforts to
and have existed as far away as the region immediately below the firmament retain or abandon the traditional concept o f celestial incorruptibility.1,2 Few
ot the fixed stars. Because comets develop from exhalations given o ff bv linked new stars to the problem o f hard or fluid heavens, perhaps because
the earth, it follows that “ if the heavens were solid, they [comets] could many scholastics either denied that new stars were real celestial phenomena
not be seen above any heaven, but all would be below the M oon, which or, if they recognized them as genuine celestial occurrences, explained them
is contrary to the common observation o f the astronomers. Therefore the as some configuration o f already existing bodies. But George de Rhodes,
heavens are fluid, so that they could be penetrated by these exhalations.” ,oX who assumed that new stars were w holly new phenomena, found it easier
Riccioli used the same argument to deny hardness to the heaven o f the to imagine such events occurring in fluid heavens. Because astronomers
Moon, although he assumed that the sphere o f fixed stars was hard because judged the distance o f new stars to be the same as those o f the fixed stars,
it seemed the only way to preserve the distances between the stars and to he found it difficult to envision how, if the orb o f the fixed stars were hard,
avoid the assignation o f a separate mover for each star.10 109
8 a new star could suddenly appear.1,3
In one o f ten arguments in support o f fluidity, Riccioli declares that the
oblique and free trajectories o f comets above the Moon, which astronomers Hi. New discoveries concerning the Moon, Mercury, Venus, Mars, and the satellites
have demonstrated,1IOwould be incompatible with solid eccentric, concen­ of Jupiter and Saturn. Other discoveries were, however, more directly rel­
tric, and epicyclic heavens. He implies that in a world o f solid orbs, epicycles evant. Melchior Cornaeus reveals the manner in which one or more o f these
would also be required to carry comets. But since comets appear only discoveries could be applied to the debate on the hardness or fluidity o f
occasionally, where would the matter come from to form a special epicycle celestial orbs. In the fourteenth century, as we saw earlier (Ch. 13, Sec.
for the occasion, and from whence would a place appear to accommodate III.7), Jean Buridan and Albert o f Saxony took opposite positions on
it? Because no such special adjustments seemed possible or plausible, Riccioli whether or not the M oon had an epicycle. Albert assumed not only that
concludes that the free trajectory o f comets demonstrates the fluidity o f the the M oon was carried around by an epicycle but that the Moon rotated
heavens, as Kepler, drawing on Tycho, argued in the fourth book o f his around its own center in the same period as the epicycle but in a contrary
Epitome of Astronomy. ' ' 1 direction. O nly in this way, he insisted, would we always see the same face
o f the Moon.
Cornaeus rejected this argument and the existence o f a lunar epicycle,
108. "Cum igitur cometae oriantur ab exhalatiombus, aut a terra elevatis aut a planetis pro-
fluentibus, si coeli solidi essent, non possent videri supra coelum aliquod, sed omnes the interior o f which would have to be either void or filled with matter. If
essent intra lunam, quod est contra communem astroiogorum observationem. Fluidi the M oon rotated around its own center as it was being carried by its
ergo sunt coeli, ut permeari possint ab ipsis exhalationibus.” Serbellonus, De caelo, disp. epicycle, the huge lunar mountains, which are greater and higher than those
i, qu. 2, art. 4, 1663, 2:25, col. 2. In my summary earlier, I omitted Serbellonus’s
assertion about the role o f planetary exhalations in comet formation. Not only was this on earth and which make the M oon’s surface rough and uneven, would
another incompatible addition to Aristotle’s comet theory (since Aristotle denied the necessarily sweep from one place to the next. As the Moon rotated, each
existence o f comets in the celestial region, no planet could have given o ff exhalations irregular lunar prominence would be carried from one place to another and
or participated in comet formation), but it makes what Serbellonus says inconsistent.
For despite the inability o f earthly exhalations to penetrate beyond the lunar orb, comets either move into a void or leave one behind, which is im possible;"4 or, it
could torm trom exhalations given o ff by planets themselves, even though their move­ the M oon’s epicycle is filled with matter, the peaks o f the lunar mountains
ments in a heaven o f hard orbs would pose serious problems. The formation o f comets would have to penetrate that matter, that is, occupy the same space with
trom matter given off by planets was a serious new theory in the seventeenth century
(see Riccioli, Almagestum novum, pars post., bk. 8, ch. 13, 1651, 57, col. 2-58, col. 2). it, which is also impossible. Insofar as his argument applies to void space,
109. Ibid., bk. 9, sec. 1, ch. 7, 244, col. 1. Cornaeus has here drawn on Aristotle’s demonstration-for a spherically
110. Although Riccioli believed in fluid heavens, the arguments presented here are drawn shaped world, which was popularized by Sacrobosco in the Middle Ages.
trom a variety ot sources. They are not necessarily his own. Indeed, a few paragraphs
earlier we saw that he thought it only probable, and not demonstrated, that comets occur A world that was not spherically shaped but had protruding angles that
above the Moon. revolved in a circle would, as Aristotle put it, “ never occupy the same
i n . M y interpretation is based on the following passage: "Quartum argumentum sumitur
a multiplici ac vago motu, seu libera et obliqua traiectione cometarum lllorum, quos 112. See Chapter 10. Section III.3.a, b; also Chapter 12, Section III. 1.
Tycho et alii censentur demonstrasse genitos et motos supra lunam fuisse. . . . Iam si celi 113. "Secundum est de novis illis stellis, quas nuper dixi apparisse saepius in coelo eandem
essent solidi, tota eorum moles cessisset in eccentncos, concentricos, et epicvclos pia- habere distantiam a terra quam habent reliquae stellae firmamenti, quod etiam cum
netarum, nec superesset materia aut locus pro epicyclis cometarum. Idcirco ex traiec- soliditate coelorum stare non potest. ’ De Rhodes. De coelo. bk. 2. disp. 2. qu. 1. sec.
tionibus cometarum fluiditatem celi demonstratum a Tychone putarunt Keplerus in 2, pt. 2, 1671, 280, cols. 1-2.
Epitome Astronomie, lib. 4, pag. 422." Riccioli, Almagestum novum, pars post., bk. 9, sec. 114. In a briefer paragraph, Riccioli seems to say the same thing; Riccioli. Alma^estum novum.
1, ch. 7, 1651, 242, col. 2-243, col. 1. pars post., bk. 9, sec. i, ch. 7, 1631, 243, col. 2, par. 21.
35« T H E C E L E S T IA L R E G IO N ARE THE H E A V E N S HARD OR F LU ID ? 359

space, but owing to the change in position o f the corners there will at one it, would smash through any hard solar heaven or orb in which the Sun
time be no body where there was body before, and there will be body again was somehow fixed. T o reinforce the argument, Amicus mentions T ych o ’s
where now there is none.” "" Because a lunar epicycle, whether void or claims about an intersection o f the orbits ot Mars and the Sun, a situation
filled with matter, seems unable to account for the rotation o f a M oon that that would make it impossible for hard orbs to exist. The satellites ofjupiter
has an irregular surface, Cornaeus rejected epicycles and assumed that the and Saturn (these were in tact the rings o f Saturn, mistaken for satellites)
Moon moves through a fluid medium. would similarly smash through any hard orbs associated with these
For scholastic authors like Cornaeus, who were compelled to reject C op­ planets.1-0
ernicus’s ecclesiastically condemned heliocentric system but who accepted
T ych o’s geoheliocentric cosm os,"6 hard, solid orbs were virtually impos­
c. Other arguments
sible. In T ych o’s scheme, which departs radically from Aristotelian cos­
mology, a number o f planetary motions are centered on bodies other than In the preceding section we saw that hard orbs were suspect because o f the
the earth. Cornaeus mentions Mercury and Venus, which m ove around the various subsystems that had emerged from the w ork ot Tycho and Galileo.
Sun as center and are therefore sometimes above and sometimes below it, It was comm only assumed that Mercury and Venus, and even Mars, moved
a state o f affairs that was based on T y ch o ’s geoheliocentric system and around the Sun as center and that the satellites ofjupiter moved around the
Galileo’s discovery o f the phases o f V enus;"7 the intersection o f the orbits latter as center; even Saturn was alleged to be the center ot two satellites
o f Mars and the Sun, so that Mars is sometimes below the Sun, and some­ that perpetually circled it. With so many centers ot motion other than the
times above it;"* and finally the tour satellites o f Jupiter are also sometimes earth, the existence o f planetary orbs appeared untenable, because it was
above Jupiter and sometimes below it, and sometimes ahead o f it and some­ assumed that as the circling bodies moved above and below the body at
times behind it. And yet all these subsystems also m ove around the earth. the center, the orb supporting the latter would be smashed, thus disrupting
It would be impossible, says Cornaeus, for these celestial bodies to be fixed the normal movements o f the heavens. Riccioli took the matter a step further
in solid, hard heavens."9 On the assumption that these arguments were by presenting an argument which declared it vain to multiply so many real
obvious to his readers, Cornaeus offers no further elaboration. and solid planetary orbs and their motions. Such a variety o f motions created
Years earlier, however, Bartholomew Amicus, who rejected fluid heavens a mutual danger o f collision and obstruction between the planets and orbs
and was a supporter o f hard planetary orbs, mentioned (De caelo, tract. 5, and also unnecessarily caused the imagination to grow weary in the con­
qu. 5, art. 2, 1626, 273, cols. 1-2) all o f the same phenomena and briefly templation o f so many allegedly real and solid epicycles, eccentrics, con-
explained w hy partisans o f fluid heavens thought the theory was compatible centrics, and epicyclic eccentrics.
with that concerning the various celestial subsystems that Tycho and Galileo As the climax o f the argument, Riccioli invokes the doctrine o f simplicity,
had identified. The phases o f Venus could not occur, he explained, unless arguing that it was unlikely that the Divine Wisdom would create a vast
the heavens were fluid, for otherwise Venus, in circling the Sun (rather than and complex machinery o f orbs to carry around a single planet like Saturn
the earth, in an otherwise geocentric universe) and m oving above and below when he could have done it so easily by the use o f a motive intelligence.
Hard orbs appear even more incongruous when one realizes that a planet
its. Ironically, Aristotle was arguing for a spherical world, whereas Cornaeus was arguing
in behalf of an irregular Moon. For further discussion, see Chapter 6, Section II and is like a point with respect to the orb that carries it - indeed it bears a
note 39. smaller ratio to its orb than any drop o f water to the ocean. W’hy construct
116. Tycho assumed that the Sun moved around an immobile earth at the center o f the world a vast orb to carry a small planet? The implication is obvious: the Divine
- just as in the Anstocelian-Ptolemaic system - but that all the other planets moved
around the Sun as the center ot their orbits. For a description o f Tycho's world system, Wisdom would have rejected hard orbs and resorted to the simpler expedient
see Thoren, 1990, ch. 8, 236-264. o f fluid h eavens."1
117. The phases of Venus are explicitly mentioned by Amicus, De caelo, tract. 5, qu. $, art.
2, 1626, 273, cols. 1-2. 120. Scholastics usually presented the best-case scenario for the opposing viewpoint, as Ami­
118. Tycho Brahe made this an integral part o f his geoheliocentric world view in opposition cus does in this paragraph. De Rhodes, De coelo, bk. 2, disp. 2, qu. 1, sec. 2, pt. 2,
to Copernicus's heliocentric system. See Thoren, 1990. 234. 1671, 280, col. 1, also cites the four satellites ofjupiter and the two attributed to Saturn
119. Cornaeus. De coelo. disp. 2, qu. 2, dub. 3. 1657, 499. Although Cornaeus does not use as evidence that “ the stars [or satellites] are moved in fluid heavens, like birds in air with
the term "hard” (durum), it is clearly implied. Riccioli, who also mentions the satellites an angel moving them through a liquid space" (ut in caelo fluido moveantur stellae, ut
o f Saturn, explains that the satellites o f Jupiter and Saturn make it unfeasible to admit aves in aere movente lllas Angelo per spatium liquidum). For Amicus’s arguments in
solid, hard orbs, because the latter would impede the motions o f the satellites. Like favor o f solid, hard orbs, see Section 2.6 of this chapter.
Cornaeus, Riccioli does not explicitly mention hard orbs - he speaks only o f “ the solidity 121. Here is the relevant text: “ Tertium argumentum. Frustra multiplicantur tot orbes reales
o f the heavens" (soliditas cAi) - but they are surely the subject o f his discussion. For, as ac solidi planetarum et motus eorum. Immo non solum trustra, sed cum periculo mutuae
we saw earlier, "solid” and “ hard” were inextricably linked in the latter part ot the collisioms et impedimenti spectata tanta vanetate motuum vel certe absque necessitate
sixteenth and the seventeenth century. cogimur imaginationem defatigare in tot realibus ac solidis epicvclis, eccentricis, con-
360 THE C E L E S T IA L RE G IO N ARE THE H EAVEN S HARD OR FLUID? 36 l
One o f the main reasons for the introduction o f epicyclic orbs in the cycle.125 N o w it is the epicycle which is assumed to move through its
ancient world was to account for variations o f planetary distances from the eccentric deferent. Under these circumstances, the inside o f the eccentric
earth. In the seventeenth century, Cornaeus rejected those orbs and ac­ will be either a plenum or a void. If void, we would have an enormous
counted for variations in planetary distance within the context o f a fluid empty space in the universe, which Cornaeus denies; moreover, motion
medium. “ If a planet sometimes approaches the earth and sometimes recedes would occur in this void, which is also denied.
from it,” he declares, “ it [the planet] is not fixed in a solid body, but it is Should the space within the eccentric deferent be a plenum, the matter
necessary that it be in a liquid and permeable body, so that at times it can will be either solid or fluid.126 If solid, or hard, the planet and the hard
approach and at times w ithdraw .” 12* matter within the eccentric deferent must interpenetrate, because as hard
Cornaeus raises an obvious objection against his own position: eccentrics bodies neither will yield to the other. But if the matter within the eccentric
and epicycles can also account for variations in distance. W hy, then, reject is fluid, one ought to say that the whole heaven is fluid, not just a part. For
hard orbs in favor o f a fluid medium? In response Cornaeus invokes tra­ if the eccentric deferent is fluid and the two eccentric orbs that enclose it -
ditional arguments that had been raised against eccentrics and epicycles since namely the eccentric thai surrounds, the eccentric deferent and the one that
the thirteenth century and were neatly summarized by Cecco d’Ascoli (see is enclosed by it - are hard and solid, how will the outermost eccentric
Ch. 13, Sec. III). Taking as his illustration the Sun, which has no epicycle communicate a motion to the inner eccentric if the two solid eccentrics are
but only an eccentric, Cornaeus argues that as the Sun moves from its separated by a fluid orb?127
farthest point frorp the earth to its closest point, it would have to move Cornaeus includes yet another important, though brief, argument in favor
through its eccentric.123 But if it does so, there must either be a vacuum o f fluidity when he asks how a vast body like the heavens could be solid
for it to m ove through or, if matter exists in the eccentric, the Sun would and hard and yet be moved with such rapidity without suffering a violent
have to penetrate that matter, either by dividing it or occupying the same disruption o f its parts and without fire arising from its intense m otion.128
place with it.124 Cornaeus then takes up the case o f a planet with an epi- Roderigo de Arriaga thought the heavens were more likely to be fluid
than solid. Neither reason nor authority suggested abandonment ot the fluid
centrids, eccentricis epicvdis. . . . Deruque incongruum videtur Divinae Sapienfiae, ut
hypothesis. Indeed, fluid heavens seemed to save diverse celestial phenom­
propter motuum unius planetae, puta Saturni, qui facillime a se vel ab intelligentia moveri
potest, moveatur tanta et tam vasta machina quanta est totum caelum cuiusque planetae, ena better than an assumption o f solidity.129
qui comparatus ad suum caelum non est nisi instar puncti et minor est quam sit gutta
respectu oceani.” Riccioli, Almagestum novum, pars post., bk. 9, sec. 1, ch. 7, 1651, 242,
col. 2, par. 1$. 2. Scholastic arguments for hard spheres
Simplicity arguments were usually invoked for any advantage they might provide to
bolster one or another side o f an argument. Because Riccioli sought to present a thorough
case for each side, he also felt an obligation, perhaps, to defend against simplidty ar­ a. The heavens conceived as a combination of hard and fluid orbs
guments that were proposed against hard orbs (ibid., pars post., bk. 9, sec. 4, ch. 33,
467, col. 2). Why would God make a world in which huge orbs had to travel at enormous A combination o f hard and soft orbs may be traced back at least to Aegidius
speeds and perhaps generate great resistances to those speeds? He could surely have Romanus in the early fourteenth century (see Ch. 13, Sec. III.9)- For Ae­
achieved the same results in a much simpler way. But these were irrelevant problems.
gidius, the eccentric deferent was like “ marrow in a bone” or “ blood in
If the celestial orbs could endure such speeds, the speeds would pose no serious problems.
Moreover, God, or the motive intelligences that move the spheres, would have no the veins,” thus associating a soft, or fluid, material with one that is much
difficulty in overcoming such potential resistances, however large they might be. Nor harder. A similar opinion, in which a solid heavens is divided into seven
indeed would our senses suffer ill effects from these great speeds, since they are regulated
by celestial intelligences. 125. Ibid., 498.
122. “ Si planeta aliquando appropinquat terrae, aliqando vero ab eadem recedit, ergo non est 126. Ibid.
in corpore solido infixus, sed necesse est ut sit in liquido et permeabili, ut aliquando 127. Here Cornaeus assumes that motion is transmitted from orb to orb.'When they specif­
possit accedere et recedere.’’ Cornaeus, De coelo, disp. 2, qu. 2, dub. 3, 1657, 497- As ically considered the motions o f the celestial spheres, scholastic natural philosophers
evidence tor distance variations, Cornaeus invoked the telescope (tubum opticum), bv assigned a separate mover to each orb, so that no orb depended on another for its motive
means o f which one could project an image o f the Sun in such a way that the Sun’s power. For a full discussion, see below. Chapter 18, Section II.
diameter would vary in size, thus indicating that its distance from the earth varied. 128. “ Vix cogitari potest quomodo corpus tam vastum et solidum tanta rapiditate moveri
Distance variations were also detectable from the observation o f eclipses. possit sine partium violentia disruptione, ac sine incendio ex nimia agitatione orto.”
123. lam not certain why Cornaeus assumes that the Sun would move through its eccentric Cornaeus, De coelo, disp. 2, qu. 2, dub. 3, 1657, 500. As part o f his defense o f the earth’s
rather than being carried by it. daily rotation, Copernicus explains that “ Ptolemy has no cause, then, to fear that the
124. “ Excentnci et epicvcli non expediunt nondum. 1: quia ut sol ex opposito Augis ex O earth and everything earthly will be disrupted by a rotation created through nature’s
veniet in / ad Augem, debet se necessano commovcre per suum excentricum. Quomodo handiwork. . . . But why does he not feel this apprehension even more for the universe,
autem hoc sit vel sine vacuo, vel sine penetratione?’’ Cornaeus, De coelo, disp. 2, qu. 2. whose motion must be the swifter, the bigger the heavens are than the earth?” Cop­
dub. 3, 1657, 497, does not provide all the details but my description seems to represent ernicus, Revolutions, bk. 1, ch. 8 [Rosen], 1978, 15-
his intent. The letters O and / are references to a figure on page 494. 129. See Arriaga, De caelo, disp. 1, sec. 3, subsec. 4, 1632, 503, col. 1, par. 41.
362 THE CELESTIAL REGION ARE THE H E A V E N S HARD OR FLUID? 363

zones or channels, was adopted in the seventeenth century by Hurtado de it as a “ very probable [valde probabilis] opinion enunciated by certain con­
Mendoza and Aversa, and was at least described by Mastrius and Bellutus. temporaries.” '14 O f the opinions they report, this may have been the one
But they cannot be said unequivocally to have combined hard and fluid they judged most plausible. Although Amicus was a defender o f hard and
parts, because they leave it unclear as to whether the hollow interior o f the solid heavens, he reported this opinion as acceptable to moderns because it
eccentric deferent o f each planet was filled with fluid matter. For those who was compatible with scriptural statements about the solidity o f the firma­
assumed it was, the hard and fluid interpretation is o f interest because it ment o f the fixed stars (probably Job 37.18, which he cites later) and also
represents an attempt to assume a degree o f fluidity while simultaneously seemed to account for the new discoveries which indicated that planets
retaining hard planetary orbs. But the most popular version that combined moved by their own motions through a fluid or airy m edium .'3''
hardness and fluidity was adopted by some o f those who abandoned all
planetary orbs except the sphere o f the fixed stars. Riccioli, who was perhaps
b. The case for solid, hard spheres
the most prominent o f this group, assumed that the sphere o f the fixed
stars was hard, whereas the region o f the planets was a fiery fluid. This Despite the inexorable, if gradual, abandonment o f hard orbs in favor o f
idea was, as he put it, “ the most celebrated contemporary opinion,” sup­ fluid heavens, a system o f hard orbs had its defenders. One o f the most
ported, according to Riccioli, by the likes o f Oviedo, Arriaga, and probably prominent was Bartholomew Amicus, some ot whose arguments were
Mastrius and Bellutus.'10 Although he recognized that this assumption was subsequently repeated by Riccioli, despite the latter’s defense o f fluid pla­
neither mathematically nor physically evident, Riccioli thought it was more netary heavens. With Amicus, scriptural arguments played a significant role,
probable than any other.'11 It seemed best suited to reconcile the numerous especially Job 37.18, which was traditionally invoked in tavor ot solid orbs.
opinions o f the Church Fathers and the theologians (that is, the scholastic Although God reproved Elihu’s discourse. Amicus interpreted this as a
doctors o f the Middle Ages). Fluidity in the region o f the planets was not moral rejection only, not one that pertained to natural things. O n the con­
only consonant with observations o f modern astronomers but required the trary, God seems to accept Elihu’s statement that the heavens are hard.'1"
least degree o f violence and the smallest number o f motions and devices. The very name firtnamentum, which applies to the heaven o f the fixed stars,
It was most compatible with the new discoveries, that is “ the phenomena implies firmness and solidity. Moreover, a solid body is required to divide
o f comets, o f Mars, Venus, and Mercury, . . . and the motions o f the sat­ the waters from the waters, since a liquid body has no proper boundaries.
ellites ot Saturn and Jupiter and o f sunspots.” ' 12 Without a solid, hard firmament to play this role, the waters would mix
As for the assumption o f a hard orb for the fixed stars, Riccioli thought with the things around them .'17 But the term frmamentum does not apply
it the best explanation to account for the unchanging distances between the only to the sphere o f the fixed stars but also to all the other heavens and
stars themselves and also the best means o f avoiding the needless multipli­ planets. After all, in Genesis 1.14 -17, which Amicus cites, God placed in
cation o f movers for the many stars.'31
134. “ Nota vero quod valde probabilis est enam quorundam recentiorum sententia ponens
The association o f a solid, hard heaven o f the fixed stars with a fluid coelum stellarum solidum in quo fixa existant astra ad motum coeli mobilia, deinde
heavens in which the planets are moved directly by intelligences without coelum aliud fluidum in quo planetae moveantur ab intelligentiis. Mastrius and Bellutus
the need o f orbs is also reported by Mastrius and Bellutus, who characterize13 2
0 [Decoelo, disp. 2. qu. 1. art. 2], 1727, 3:490, col. 2, par. 3S. Earlier we saw that Mastrius
and Bellutus described another opinion in which the heavens were divided into seven
zones or channels (see Ch. 13, Sec. III.9 and n. 112). The opinion they Favored is left
130. “ Et nunc celeberrima opinio.” Riccioli, Almagcstum novum, pars post., bk. 9, sec 1. ch. uncertain.
7 , i<Mi. 240, col. 1. For the names ot its supporters, see 240, cols. 1-2. This “ most 13 s. “ Secunda opinio est dicentium caelos planetarios esse ex materia liquida quia tacilius
celebrated opinion was the tilth ot those that Riccioli categorized under the fluidity ot motus et apparennae planetarum salvantur caelum; vero stellarum tixarum esse solidum
the heavens. in quo stellae sunt Fixe ut nodi in tabula, unde non per se sed ad motum orbium moventur.
131. “ Probabilius multo est, licet nondum mathematice aut phvsice evidens, caelum tixarum Hanc significant dicentes Stellas in firmamento tixas esse; planetas vero per aerem vagan
solidum esse, planetarum autem fluidum.” Ibid., 244, col. 1. vario motu. . . .
132. “ Constat id arguments et responsiombus utnmque hactenus adductis: hac emm dis­ Hec tamen opinio potest a recentioribus accipi quia ex asserentibus liquiditatem ce-
tinction turn probabilitatis ab evidentia turn caeli tixarum a caeli planetarum conciliantur lorum est omnium probabihssimam satisfied scripturae tribuenti Firmamento solidita-
plurime opimones patrum ac doctorum inter se et cum astronomorum recentiorum tem, quod est celum stellatum tixarum. et sansFaciunt novis apparentiis dum planetas
observatiombus minorique violentia aut multiplicitate motuum ac machinarum; minon naoveri per se propriis motibus per orbem atterunt." Amicus, De caelo, tract. 5, qu. 5,
quoque penculo repugnantiae phvsice inter motus tarn vanos planetarum explicantur art. 2, 1626, 274, col. 2. In this report. Amicus seems to envision the planets as moving
phenomena cometarum, Martis. Veneris ac M ercuni. . . et motus comitum Saturni et through a fluid medium that is located within an orb.
Iovis et macularum solarium.” Ibid. This was essentially Arriaga's opinion (see Section 136. After citing the passage. Amicus says: “ Neque obstat quod illud sit dictum ab Elihu
VIII. 1 c o f this chapter-!. cuius discursus tint a Deo reprobatus c.38. nam ibi t'uit reprobatus discursus moralis
133. “ Si sphaera tixarum solida ponatur, prompdus redditur ratio cui servent perpetuo eam- quo Job sanctitatem innocentiam accusabat. Sed quoad naturalia ibi ducta a nullo sapiente
dem inter se distantiam; nec multiplicandi erunt innumerabiles motores tixarum.” Ric­ reprobantur, sed potius recipiuntur.” Amicus, ibid., art. 3, 278, col. 1.
cioli, Alma^cstum novum, pars post., bk. 9, sec. 1, ch. 7, i As i , 244, col. 1. 137. Ibid., 278. cols. 1-2.
364 THE CELESTIAL REGIO N ARE THE HEAVENS H A RD OR FLUID? 365

that very firmament the luminaries he created on the fourth day. Planets If the celestial substance were really fluid, the enormous velocities o f the
and stars are all part o f the firm am ent.'3S gigantic celestial bodies that move through it would seem o f necessity to
Reason also tells us that solid, hard orbs are needed to carry the planets produce a loud noise, especially at the point o f im pact.'43 Although Amicus
perpetually at great velocities. Without them, the celestial bodies would fails to draw the inference, it is obvious: because we hear no such sound,
self-destruct and fail to preserve the order and constancy o f the heavens 139 the heavens cannot be o f a fluid nature.'44 On a more positive note, Amicus
Amicus demonstrates this with three brief arguments: (1) If the planets were declares that solid, interconnected, and interrelated orbs confer a greater
not embedded in solid orbs, one intelligence would be required for each nobility and system on the heavens than would be the case with stars and
planet and each fixed star; so great would be the number o f intelligences planets m oving through a fluid medium as fish move through the sea.'43
required that a needless multiplication o f entities would result.140 (2) A Amicus also argued that the attribution o f liquidity to the firmament was
mover for each planet would be a much less effective way o f preserving contrary to common sense. But though he was a staunch advocate o f solid
uniformity o f celestial motion, just as it would be if all the stars and planets orbs, Amicus conceded that solidity was originally an unnatural state for
were moved only by a single mover. For in the latter situation, the intel­ the heavens. When God created the heaven on the first day, he apparently
ligence, or angelic intellect, would, because o f its finitude, be less able to produced a fluid, watery heaven. The true nature o f the firmament was
attend to individual stars and planets.'41 Finally, (3) because angels move thus fluid. O n the second day, however, God, in dividing the heaven o f
bodies only when they are actually in touch with them, it follows that if the first day, made the solid firmament, which is really the fluid heaven o f
angels moved the stars and planets directly, each angel would have to be the first day made hard and solid. The solidity o f the firmament is therefore
moved right along with its ow n celestial body in order to assist it as it an accidental property o f the heavens, secondary to its true and original
moves along, a situation that is avoided if the planets are embedded in sohd fluid nature.'46 Amicus observes that there are those who believe that the
orbs.142 heavens are naturally fluid and remained that way and there are others who
hold that the heavens were originally fluid but were made unnaturally solid
138. “Secunda conclusio soliditatem. quam probavimus convenire firmamento, probabile est and hard. Because the fluidity o f the heavens seems natural in both theories,
convenire omnibus caelis etiam planetarum." Ibid., 279, col. 2.
139- “ Secundo probo racione: nam orbes sunt ordinau ad deferenda svdera perpetuo et or­
Amicus concludes “ that from authority, from the motions o f new stars,
dinate. At hie finis exigit soliditatem orbium quia alioquin per tantam velocitatem motus and from similar things, which [Christopher] Scheiner reports, it is suffi­
facile corpora dissparentur atque adeo non posset servan tanta constants et ordinatio ciently probable that the heavens are fluid. But I do not follow this [opinion],
motuum et corporum motorum.” Ibid.
NO- Turn quia si sydera pe se ct non infixa orbibus moverentur magnus exigeretur intel-
ligentiarum numerus ad tot Stellas presertim tixas ordinate movendas unde sine neces­ bodv to which it is assigned. For more on Riccioli’s ideas on the application o f impressed
sitate sutficienti multiplicarentur entia. Ibid. Riccioli s eighth argument in favor of solid forces to celestial bodies, see Chapter 18, Section II.6.a and note 226.
orbs is similar. Si non concedantur orbes sohdi, oportebit ad movenda corpora celestia 143. “ Conf. quia mirum est ex tanta velocitate motuum corporum liquidorum tarn ingentium
multiphcare innumerabiles intelhgentias, tot mmirum, inquit Tannerus, quot sunt stellae non gigni sonum adeo ingentem ut ad nos pervemat, nam sonus gignitur ex collisione
fixae, quotque maculae soils. Riccioli, Almagestum novum, pars post., bk. 9, sec. 1, ch. corporum ad acrem, id est, corpus liquidum. Amicus, ibid.
7. 1651. 24.1. col. 2. 144. Riccioli, Almagestum novum, pars post., bk. 9, sec. 1, ch. 7, 1651, 241, cols. 1—2, describes
141- Turn quia ex motibus factis a tot diversis motoribus non posset servari tanta uniformitas the same argument mentioning that the sounds should be akin to those hissing or
quia non possent semper attendere ad servandam tantam vel tantam velocitatem, et alias whistling sounds that emanate from stones launched in the air by slinging machines. He
circumstantias, ex quibus pendet uniformitas motus. Et hoc idem probat si omnes stellae also cites counterarguments trom Tycho Brahe, Christoph Rothmann, and Franciscus
moverentur ab eodem. nam licet mtellectus Angelicus comprehendat naturas rerum de Oviedo. Thus Rothmann denied that such sounds could reach our ears, because ot
corporalium, adhuc tamen ob sui fimtatem minus potest attendere ad singula, dum the great distances and the rarity o f the celestial ether. Oviedo’s response was predicated
plunbus lntendit.” Amicus, De caelo, tract. 5, qu. 5, art. 3. 1626, 280, col. 1. on the familiar analogy between the movement ot fish m water and planets in the heavens.
[*G- Turn quia cum Angeli non moveant corpora nisi sint sibi praesentia. necessano oporteret Just as there is no sound in the water itself when fish swim through it, so also there is
ipsos movcri cum ipsis stellis, assistendo lilis ut moveant, quae omnia vitantur si moven no sound in the fluid heavens as the planets move through them. Moreover, defenders
dicamus in orbibius infixas.” Ibid. Amicus probably derived the argument about the o f hard orbs ought to be asked why fire and air do not produce audible sounds as a
need for angels to be present where they act from Thomas Aquinas, who denied action result o f the circulation o f the lunar heaven.
at a distance even for spiritual creatures, including God (see Grant, 1981a, 146). In his “ Conf. secundo nam quo corpora sunt supenora eo magis sunt nobihora et maion
category of arguments in defense of hard orbs, Riccioli. Almagestum novum, pars post., quodam artificio ornata. At hoc artificiuin magis apparet ponendo multos orbes turn
bk. 9, sec. 1, ch. 7, 1651, 241, col. 2, presents the same argument. As a counterargument, mobiles inter se connexos et ordinate m otos. . . quam si ponatur unum liquidum per
he declares that an intelligence need not travel around with the body it moves but could quod stellae discurrant ut pisces per mare.” Amicus, De caelo, tract. 5, qu. 5, art. 3,
rather remain in a particular place and impress enough impetus into a star or planet to 1626, 280, col. 1.
carry it around tor one revolution. When the point at which the impetus had been “ Firmamentum secundo die productum sola sohditate ditfert ab eodem producto initio,
injected arrives again at the same place, the intelligence impresses the same quantity of sed soliditas, cum sit accidens, non variat naturam rerum, ergo neque naturam tirma-
impetus to carry the body around tor another revolution, and so on. But Riccioli also menti. Si prius erat liquidum ex natura. similiter ent natura liquidum sub soliditate. Hec
insists that no difficulties would arise if an intelligence moved around with the celestial autem variatio in caelo facta est ob bonum universi. Ibid., art. 4, 281, col. 1.
366 THE CELESTIAL REGION ARE THE H E A V E N S HARD OR FLUID? 367

nor do I retreat from ancient opinion without an urgent reason and [also] cites one that was intended to show a fatal flaw in the theory o f fluid heavens.
because solidity [and hardness] conform more to Scripture to which every On the assumption that the planets move through fluid heavens that could
human intelligence is subjected.” '47 Although Amicus thought the fluidity have no void spaces and in which bodies cannot penetrate one another,
o f the heavens improbable on scriptural grounds, he conceded that the Riccioli explains that a planet m oving through such a fluid would cause
opinions drawn from scriptural texts and Church Fathers were not so clear either the whole fluid, or a part o f it, to undulate. This would occur by the
or obvious in support o f the solidity o f the heavens. M oreover many learned impact o f the planet’s forward motion, which would condense the fluid in
contemporary theologians, philosophers, and astronomers (he calls them front o f it. As this occurs, the matter behind the planet would necessarily
“ mathematicians” ) thought they were fluid. For these reasons, Amicus de­ rarefy in order to prevent formation o f a vacuum in the places that the
clares that despite the improbability o f fluid heavens, it was by no means planet has just vacated. But condensation and rarefaction are alien to the
rash to uphold this theory.14* heavens and are signs o f corruptibility. Fluid heavens would thus produce
Giovanni Baptista Riccioli represents another significant seventeenth- impossible consequences.
century source for arguments favoring celestial solidity and hardness. Al­ Riccioli also reports an argument in which hard orbs are judged better
though he rejected heavens filled with solid orbs. Riccioli, like many other for explaining the occurrence o f a plurality o f simultaneous, and even con­
scholastic authors, sought to present a balanced account in the dispute over trary, motions. Indeed, it is impossible for the same body to move with
fluid or solid heavens. Just as he did in his section defending fluid heavens, several motions simultaneously unless it achieves this with a motion o f its
he compiled arguments in favor o f total or partial solidity and hardness and ow n (per se) in one direction while the one or more remaining motions are
often included a common rebuttal o f each argument. Riccioli reiterated what produced by the motion o f the solid body to which it is affixed or in which
was probably obvious to everyone by his day: solidity signifies not only it is embedded.1' 3
three-dimensionality but also has an associated meaning o f hardness as op­
posed to softness.14149 As he did in his presentation favoring fluid heavens,
18
7 c. The new discoveries and solid orbs
Riccioli cites arguments from authority in support o f celestial solidity, draw­ T o defend solid, hard orbs, it was essential to deny that the new discoveries
ing upon Holy Scripture (including Job 37.18, the most frequently men­ implied fluid heavens and/or to deny or cast doubt on the new discoveries
tioned), the Church Fathers,1'0 and Aristotle. Thus we are told that in De
caelo (bk. 2, ch. 7) Aristotle declares that a heaven and the star or planet 152. “ Quartum argumentum: Si celum in quo moventur sidera esset fluidum et non admittatur
vacuum aut penetratio corporum, sequeretur ad motum sideris vel totum fluctuare celum,
that is part ot it are made o f the same material. But a star or planet is a vel partem celi a sidere impulsam condensan, partem vero a sidere destitutam raretien.
solid body; therefore, so is the heaven or orb that carries it.1' 1 At condensatio et rarefactio repugnant caelo et sunt indicium corruptibilitatis." Ibid. To
Leaving authority and moving on to more substantive arguments, Riccioli this argument, Riccioli presents a number o f replies. Some argue that in the celestial
region mutual penetration o f bodies is possible, or that rarefaction and condensation are
possible in incorruptible heavens; and some insist that the heavens are corruptible and
147. “ Ex quibus puto satis probabiie esse caelos esse fluidos ex auctoritate, et monbus novarum that therefore condensation and rarefaction can occur. During the entire period covered
steilarum et simihbus. quae affert Schemer. Sed earn non sequor. ne recedam ab antiquata bv this study, fear o f the consequences that bodies and celestial matter might interpe­
opimone sine ranone urgente et quia soiiditas est magis contormis senpturae cui omnis netrate and that celestial matter might rarefy and condense frequently compelled de­
humana intelligentia subdidebet.“ Ibid.. 282. col.i. fenders o f celestial orbs to explain how such dire consequences could be avoided (see
148. “ Ego vero in hac diversitate opinionem asserennum caelum esse liquidum existimo esse Ch. 13, Sec. Ill, especially 1-3 and n. 59).
quidem improbable, non tamen temerariam. Nam senpturae loca et Patrum testimonia 133. “ Sextum argumentum indicatum ab Anstotele et incuicatum a Pereiro sumitur a mul-
non in dare soliditatem caelorum expnmunt. ut interpretationem non admittant ut patet tiplicitate inotuum. Impossible enim est idem corpus moveri plunbus motibus et quidem
ex iis quae adversani adducunt. Idque confirmo nam nostre aetate multi sunt ex Theo- contrariis. quomodo constat moveri planetas. immo et hxas, nisi uno motu per se
logis, Philosophis, et Mathematicis, multae eruditionis, qui liquiditatem caelo convenire moveantur in unam plagam. reliquis autem moveantur ad motum corporis solidi. cm
nituntur probare quos non est aequum temeritans censura notan.” Ibid., art. 3, 281, sint affixa aut insidentia.” Ibid., 241, col. 2. As a counterargument, Riccioli declares
col. i. that even if another bodv were required to account for all the motions o f a single planet,
149. Riccioli, Alma^estum norum, pars post., bk. 9, sec. t, ch. 7, 1651, 240. col. 2-242, col. it would not have to be solid and hard. After all, fish are carried downstream by a
1. tor the arguments, and 238. col. 2, for the linkage between solidity and hardness. rapidly moving river while they simultaneously attempt to move upstream in the op­
130. Ibid., 240, col. 2-241, col. 1. posite direction. But Riccioli reports further that Clavius had countered this argument
i s i . “ Tertium argumentum est Aristotelis lib.2 De celo, cap.7, ubi ait congruum esse ration! by observing that if the planets were really moved freely through a fluid medium like
ut caelum sit ex eodem corpore cuius est sidus; sidus autem quodlibet esse solidum. ergo fish and birds, their motions would be just as indeterminate and uncertain. It would be
et celi corpus ex quo est." Ibid., 241, col. 1. For the passage in Aristotle, see De caelo as if they did not know that they were being moved in a fluid medium by an intelligence.
2.7.289a. 11 —12. In the counterargument, Riccioli concedes the identity o f the matter of Hence we could have no certain knowledge o f planetary motions. (“ Aliter hoc argu­
the planet and its orb or heaven but insists that they differ in other wavs, for otherwise, mentum proponit Clavius in cap. 4, sphere pag. 449. ait enim si moverentur ut pisces
it would follow that because a planet is luminous and opaque, so also would the whole et aves in fluido, liberum ac nimis vagum tore planetarum motum et sic nullam certain
heaven be luminous and opaque. But this is false. Almagestum novum, pars post., bk. 9. fore scientiam de ipsorum motibus. quasi vero nequeant ab intelligentia moveri in fluido,
sec. 1, ch. 7. 1631, 241. col. 1. servatis tamen legibus motuum." Ibid.)
368 THE CELESTIAL REGION ARE THE HEAVENS HARD OR F LU ID ? 369

themselves. Amicus did both. Following Tanner, Amicus argues that the the fluid heavens between two hard surfaces, with the planets and stars
new phenomena indicate that certain planets - presumably Mercury and distributed at various altitudes between th em .'57
Venus - do not have proper and distinct orbs that surround the earth but In a certain sense, the controversy had reached a stalemate. For although
are carried in epicycles around the Sun and actually lie within the Sun’s orb solid orbs seemed incapable o f explaining the occurrence o f celestial comets
Thus did Amicus, and others, accept that part o f T y ch o ’s system which or new stars, partisans o f solid orbs could, as we have seen, either deny
made the Sun, rather than the earth, the center o f the orbits o f Mercurv that such phenomena were celestial or invoke normally invisible celestial
and Venus. Indeed, the result was a variation o f traditional geocentric cos­ bodies that are carried in epicycles within larger spherical complexes and
m ology that resembles the medieval opinion reported in the fourteenth that somehow cluster together to produce a visible body. O n this approach,
century by Jean Buridan.'54 And just as Buridan thought this configuration change is accidental rather than substantial and involves a mere rearrange­
probable, so did Amicus assume that the epicycles - he makes no mention ment o f previously existing bodies. As for the various subsystems with
o f eccentrics - which carry Mercury and Venus around the Sun do so centers other than the earth, these were usually explained by the assumption
without any penetration o f one orb by another and without any crashing o f epicycles for the satellites themselves or for Mercury, Venus, and even
o f orbs.15
155
4 Mars. Sunspots were also explained in a similar manner. Although in ret­
rospect such constructions seem o f an ad hoc nature, they enabled a steadily
diminishing group o f scholastics to hang on to the chief elements o f the
» IX . T h e d iv e rsity o f o p in ion old system despite the ovewhelm ing challenge that confronted them.
Within the broad categories o f hard and soft, a wide variety o f interpre­
What do all these diverse and often conflicting opinions signify? Probably tations was formulated, with no decisive way to choose among them. The
an inability to determine convincingly the operational structure o f the heav­ seeming advantage o f the theory o f hard and solid eccentrics and epicycles
ens. Numerous opinions and variations on those opinions were inevitable, was that the planet was carried around within a hard epicycle. Although
because scholastic natural philosophers did not and could not know the true there were grave problems about the place o f the planet within the epicycle
nature o f the heavens. There was much room for disagreement about the itself and how the planet would relate to either a vacuum or some kind o f
hardness or fluidity o f the celestial region. Even some o f those who adopted matter within the epicycle, the planet was at least fixed within the epicycle.
a fluid medium for the planetary region opted for a hard orb for the fixed But if each planet seemed to be carried within its hard epicycle, the problem
stars, not only because it seemed more economical to have one mover for o f motion was certainly not solved; rather, it was removed one step to the
all the stars rather than one mover for each star, but perhaps also because orb itself. What caused the orbs themselves to move is a problem that I
it seemed more fitting that the whole o f the cosmos be enclosed by a firm, shall consider later (Ch. 18).
hard surface to keep it all together and prevent its dissipation into the region A nd yet the fluid theory o f the heavens emerged as the most appropriate
beyond.156 Indeed, Galileo himself seems to have thought it best to enclose interpretation o f the new celestial phenomena. It triumphed because it
seemed more congruent with those phenomena and made fewer incredulous
154- See Chapter 13, Section IV. 1, for a discussion and an English translation ofBuridan’s
text (n. 137 for the Latin text).
demands on reason. Fluidity did not, however, triumph because o f any
155. Ad tertium ductum ex novis phaenomenis, resp. Tannerus disp. 6, q. 3, d. 3, pa;. 6, overwhelm ing and certain arguments. Indeed, those who abandoned hard
ex horum planetarum motibus id solum consequi non habere proprios et distinctos orbes orbs in favor o f fluid heavens had to confront the problem o f planetary
terrain ambientes, sed solum epicyclos in orbe solari solem ambientes per quos planetae
motion directly. What enabled a planet to move in its orbit like fish in the
circa solem feruntur sine ulla vel penetratione, vel fractura orbis, quod stat cum caeli
soliditate et incorruptibilitate. sed his obstat communis sensus philosophorum et theo- water or birds in the air, as the popular analogies expressed it? For those
logorum, qui illis planetis tribuunt proprios orbes mobiles circa centrum mundi. Sed who not only assumed fluid planetary spheres but also a fluid zone for the
hoc non putant absurdum.” Amicus, De caelo, tract. 5, qu. 5, art. 3, 1626, 285, col. 1.
fixed stars, there was the additional problem o f assigning a motive cause
Amicus did not agree with those philosophers and theologians who insisted that all orbs
must have the earth as center. The new astronomical observations showed that Mercury to each o f the more than one thousand visible stars. For no longer could
and Venus moved around the Sun in proper epicycles. (“ At cum observatum sit per they rely on a single hard orb to carry around the fixed stars that had been
novas observationes illos planetas non moveri circa terram, sed circa solem; ideo negandi
previously imagined as fixed in their hard sphere like knots in a piece o f
sunt illis proprii orbes et ob id conceduntur propni epicvcli.“ Ibid.) Serbellonus, De
caelo, disp. 1. qu. 2, art. 4, 1663, 2:26, cols. 1-2, reports that not only are Venus and w ood. Although he recognized that if the firmament were solid, only one
Mercury contained in the solar orb in proper epicycles but so also is Mars. He rejects
the existence of solid eccentrics and epicycles because division, penetration, and fracture 157. Galileo, Dialogue, Third Day [Drake], 1962, 325-326. Although the words are put into
ot orbs would inevitably occur. the mouth o f Simplicio, they seem to represent Galileo's opinion. As early as March
[56. According to Van Helden (1985, 63), “ Kepler continued to believe,” to the very end ot 23, 1615 (in a letter to Monsignor Dini), Galileo had already rejected the real existence
his life, “ that the fixed stars were arranged in a spherical shell.” o f “ solid, material, and distinct orbs” (see Finocchiaro, 1989, 61-62).
370 THE CELESTIAL REGION

mover would be required to carry all the stars simultaneously, Melchior 15


Cornaeus preferred to believe that God did not create hard orbs but rather
assigned one angel to each star, o f which there were more than a thousand.
Alter all, God was not destitute o f angels, and a star was not so small that
it did not deserve its own m otive angel.,vS The immobile orb
Thus, where Kepler had proposed a causal, physical mechanism based
on magnetic forces to account for the motions o f orb-free planets in his
Astronomia nova (1609) and Epitome astronomiae Copenicanae (1617—1620),159
o f the cosmos:
scholastic natural philosophers, who assumed, as did Cornaeus, fluid heav­
ens, relied heavily upon angels or intelligences as celestial movers. But unlike
the empyrean heaven
Cornaeus, as we shall see in Chapter 18, some associated impressed forces
with angelic movers and thus tended to make the latter more mechanical
than spiritual. In so doing, they joined Kepler and others in a quest for
more impersonal forces to explain the motion o f orb-less celestial bodies. Thus far we have considered only mobile orbs, all ot which had astronomical
Until the theory o f universal gravitation in N ew ton’s Principia settled the functions and at least two o f which - the eighth (the firmament) and ninth
matter once and for all, there were only ad hoc theories, which w'ere no (the crystalline orb) - also had biblical sanction.1 In the chapter on creation,
more convincing than the causal explanations invoked for celestial orbs. we saw that in commentaries on Genesis the heaven, or orb, created on the
Nonetheless, the cumulative evidence inclined strongly and suggestively first day was often called the “ empyrean,” ' even though the latter is not
toward fluidity, and more and more scholastics embraced it as the most explicitly mentioned anywhere in the Bible. Its existence was derived trom
plausible alternative. faith, not rational argument. Inferences about it tashioned an invisible,
immobile orb that enclosed the world, a place that was conceived by many
158. Cornaeus, De coelo, disp. 2, qu. 2, dub. 3, 1657, 500, first raises an objection against
himself: "Si firmamentum non est solidum. ergo singulis asms assignandus est angelus as the “ first and highest heaven, the place o f angels, the region and dwelling
motor, qui per liquidum eonducat et certo itinera dirigat. Atqui si firmamentum sta- place o f blessed men’” or as “ the dwelling place o f God and the elect.” 4 If
tuamus solidum. unus sufficiet pro omnibus" and then replies: "Concedo sequel. Neque
the number o f mobile orbs is ?i, the empyrean orb was always numbered
tarn parva res est Stella ut angeli custodiam non mereatur, neque tarn inops angelorum
est Deus ut pro omnibus et singulis stellis non sit ei sufficiens eorum copia." We shall n + 1. In the most popular, ten-orb mobile system, it was the eleventh
learn more in Chapter iS about causal factors in celestial motion. and final orb.''
i>9- Kepler relied on two forces. According to Kovre (1973, 323). he assumed a rotation of
the Sun, which “ sends out into space (in the plane of the ecliptic) a motive whirlpool
which carries the planets round and impresses on them a circular motion round the Sun;
at the same time the planetary magnets, in accordance with a mechanism which has
been fully described above, causes the planets to approach and recede from the Sun. As I. Features and p roperties o f the em p yrean heaven
a result o f being subjected to this two-fold influence, the planets do not describe circles
in the sky, but describe ellipses having the Sun at one o f their foci.” According to Thomas Aquinas, the empyrean heaven had been recognized
much earlier by Venerable Bede and Walafnd Strabo/’ Without using the

1. The "crystalline orb" is biblical only when :t is interpreted as the congealed form ot the
waters above the firmament. As we saw earlier iCh. 13, Sec. IV.2.b). the tenth orb might
also be included within the concept o f a crystalline orb.
2. For a description o f some o f its salient features, see Chapter 6, Sections II and III.2.1
and g.
j "Coelum empvroeuin est primunr et sununum coelum. locus angelorum, regio et habi-
taculum hommum beatorum.” Bartholomew the Englishman, De rerum proprietatihus. bk.
8. ch. 4 ("De coelo empvreo"). 1601. 379-380. Although God is everywhere, Bartholomew
says that he is especially in the empyrean heaven.
4. "Coelum empirreum habitaculum Dei et omnium electorum" is the description found in
the figure o f the world that appears on column 6 ot Peter Apian's Cosmooraphiais liber
(Landshut: Johann Weyssenburger for Petrus Apianus, 1524), reproduced in this volume
as Figure 9.
5. For this eleven orb system, see Figure 9 -
6 . Sutnttiu tlteolo^toe, pt. 1, qu. 66 . art. 3, iqfw. 4*2- Although Saint Basil s role was incon­
sequential, Thomas included him as a third person involved in the development ot the

371
372 THE CELESTIAL REGION THE IM M O B IL E ORB: THE E M P Y R E A N 373

word empyreum, Venerable Bede had indeed, in the eighth century, distin­ to be a b o d y ,14 the empyrean heaven was the most subtle o f all bodies and
guished an immobile heaven created on the first day from the mobile, contained within itself the purest light in the universe. Albertus Magnus
observable celestial bodies created later.7 Without specifying immobilitv, envisioned it as formed from fire, the most noble o f the simple elemental
Alcuin did much the same thing in the ninth century.8 bodies, but Thomas Aquinas denied it any connection with the four elements
But the empyrean sphere did not emerge as a distinct entity called the and insisted rather that it was composed o f pure ether, the fifth element in
caelum empyreum until the twelfth century, when Anselm o f Laon, Peter Aristotle’s cosm ology.'5 Despite the empyrean’s purity and splendor, it was
Lombard, Hugh o f Saint Victor, and Roland Bandinelli (the future Pope invisible16 and its effects were imperceptible. Like the other celestial spheres,
Alexander III) furnished brief descriptions that gained acceptance in sub­ the empyrean heaven was thought incorruptible; unlike them, however, it
sequent scholastic literature. According to a description by Anselm o f Laon was always assumed immobile.
(d. 1 1 17) in the Glossa ordinaria,9 the empyrean sphere was understood as In the fourteenth century, Thomas o f Strasbourg (fl. 1345) ([Sentences,
“ fiery or intellectual, which is so called not by virtue o f its burning [or bk. 2, dist. 2, qu. 2], 1564, 134G col. 1-134V, col. 1) gathered much that
heat] [ardor] but from its brilliance [splendor], since it is immediately filled had been elaborated in the preceding century and presented a thorough
with angels.” 10 Not only did Peter Lombard, in his famous Sentences, quote discussion o f the properties o f the empyrean sphere, to which he assigned
the words o f Anselm, but he also identified the empyrean sphere with the four basic attributes. It is the most lucid, or light-filled, sphere. As the first
invisible heaven created on the first day and thus distinguished it from the and noblest o f celestial bodies, the empyrean ought to possess the noblest
heaven created on ihe second day and made visible on the fourth day.11 corporeal quality, which is light. Moreover, as the noblest o f bodies, it
Indeed, Peter believed that the empyrean heaven was created simultaneously should be independent o f all other celestial bodies. Because the latter receive
with the angels and all corporeal things.12 their light solely from the Sun, the empyrean should receive its light directly
During the thirteenth century, all the great theologians — William o f from God. Although it was filled with light, the empyrean sphere trans­
Auvergne, Alexander o f Hales, Albertus Magnus, Saint Bonaventure, Duns mitted no light to the celestial and terrestrial regions below. Empyrean light
Scotus, Richard o f Middleton, and Thomas Aquinas — had come to accept was effectively blocked by the nontransparent nature o f the eighth sphere
the empyrean heaven. B y then it had become the dwelling place o f God o f the fixed stars.
and the angels, as well as the abode o f the blessed.13 Although conceived B y contrast - and this is its second property - the empyrean sphere was
transparent and rarefied, made so to enhance the pleasure o f the blessed, so
concept o f an empyrean sphere. For a brief but useful account o f the empyrean sphere that each inhabitant could see friends in the same state.
as it developed in the late Middle Ages, see the article on it in the Dictionnaire de theologie Its third property was incorruptibility, which was inferred from Aris­
catholique. vol. 2, pt. 2. cols. 2503—2508. The earlier history is described by Maurach,
1968. An important collection ot Latin descriptions o f the empyrean sphere from Bede totle’s general attribution o f incorruptibility to the other planets. Because
to Thomas Aquinas appears in Litt, 1963, 255-258, n. 1. Duhem included a brief account the empyrean was nobler than all other celestial bodies, it must also be
in Le Systeme, 1913—1959, 7:197-202, o f which pages 197-200 have been translated in incorruptible.
Capek, 1976, 4.3-45. See also Grant, 1978a, 275-276. For an account o f the fate o f the
empyrean heaven in the first half o f the seventeenth century, see Donahue, 1972, 223-
Finally, Thomas o f Strasbourg argues that the empyrean heaven must be
259. immobile. Immobility is the only appropriate state for the blessed, who are
7. Bede [Genesis], 1967, 4, lines 35—41. The term empyreum had already been applied to the themselves in a perfect state o f rest. Although Aristotle denied the existence
extramundane region by Martianus Capella in the fifth century. In Campanus o f Novara,
Theorica planetarum. 1971, 393. n. 52, Benjamin and Toomer cite the Latin text from
o f immobile spheres, Thomas declares that this has no validity for Chris­
Dick's edition ot The Marriage of Mercury and Philology (De nuptiis Philologiae et Mercurii tians, “ who assume a certain body above the first movable body [pritnum
lihri VIIF), bk. 2. line 200. p. 76. mobile] itself, [a body] that is absolutely independent o f the first movable
8. For the passage, see Litt, 1963. 257.
9. During the Middle Ages, the Glossa was falsely ascribed to Walafrid Strabo (this attribution sphere, as is the empyrean heaven itself.” The empyreaft heaven is not,
is repeated by Benjamin and Toomer (see Campanus o f Novara, Theorica planetarum,
1971, 393 - n- 52)- On the assignation o f the Glossa ordinaria to Anselm, see Thomas the empyrean heaven with one o f the astronomical orbs, namely the ninth, or crystalline,
Aquinas, Summa theologiae, 1967, 10:40. n. 5. sphere. In 1241 and then officially in 1244. the bishop o f Pans condemned those who
10. My translation is from the Latin text quoted in Campanus o f Novara, ibid. The translation located the Blessed Virgin and the glorified soul in the ninth, or crystalline sphere, instead
first appeared in Grant. 1978a. 275—276. This assertion was frequently repeated (see, for o f in the empyrean heaven, or tenth sphere. Indeed. Michael Scot explains that the
example, Vincent of Beauvais. Speculum naturale, bk. 3, ch. 88. 1624. col. 220). empyrean heaven was introduced to serve as a moving cause prior to the first movable
11. The brief biblical description o f a seemingly distinct heaven created on each o f the first sphere, or primtim mobile. For all this, see Vescovini Federici s discussion in Peter o f Abano,
two days formed the basis for belief in an empyrean heaven. Because the heaven ot the Lucidator, 1988, 200—201. It appears that the ecclesiastical authorities wished to have a
second day was clearly intended as the firmament, only the vaguely described heaven ot place for the blessed that was distinct from any sphere that had an astronomical function.
the first day remained as a viable candidate for the empyrean. 14. See Alexander o f Hales, Summa theologica, inquis. 3, tract. 2. qu. 2, tit. 1, memb. 1, art.
12. Peter Lombard [Sente; es, bk. 2, dist. 2, ch. 5], 1971, 340; see also Litt, 1963, 256-257- 2. 1928, 2:329 and Dictionnaire de theologie catholique. vol. 2, pt. 2, col. 2506.
n. 1. 15. Dictionnaire de theologie catholique, vol. 2, pt. 2. cols. 2506—2507.
13. During the first half o f the thirteenth century, there was apparently an attempt to identity 16. See Chapter 16, Section I and note 7.
3 74 THE CELEST IA L REGION THE IM M OBILE ORB: THE E M P Y R E A N 375

however, inherently immobile, because, like any celestial sphere, God could Aristotle when he departs from Scripture,1" and Buridan subsequently dis­
move it if he wished. But the empyrean heaven is called immobile, because misses the arguments just described. As the noblest body, the empyrean
universal order does not require its motion and because God does not subject sphere does not require action or motion. And although it could indeed be
it to the power o f any other creature that could move it. moved supernaturally by God, the empyrean sphere has no inclination -
that is, no potentiality - for motion and therefore would not be eternally
frustrated by its failure to achieve it.
Despite his seeming defense o f the empyrean sphere on theological
grounds, Buridan’s commitment to it is doubtful. At the end ot the question,
II. A rgu m en ts fo r and against an im m o b ile sphere
he adds a section for those who wish to defend the Aristotelian position
Although an immobile empyrean orb was widely accepted on theological, and promptly offers guidance by refuting five arguments in favor o f the
rather than cosmological, grounds, its alleged immobility clashed with Ar­ empyrean sphere that he had previously presented.'0 One ot the favorable
istotelian natural philosophy, which cast serious doubt on such a possibility. arguments - the fourth - held that the earth was divided into two halves
It was, therefore, not unusual for scholastics to inquire whether an immobile by the equatorial circle. The half toward the antarctic pole is uninhabitable.
orb could exist beyond all the mobile orbs, as Jean Buridan makes evident The other half, toward the arctic pole, is divided into two quarters, one ot
in his discussion o f the question "W hether a resting or immobile heaven which is habitable, the other uninhabitable. Since the earth is ruled by the
should be assumed above the heavens that m o v e .'"7 heavens, it is essential that the heavens arrange things so that one ot these
two northern quarters is habitable and the other uninhabitable and covered
with water. But this arrangement cannot be caused by mobile heavens,

i. The arguments against


because the same parts o f the heaven and the same planets [and stars] are turned
First, Aristotle held that every natural body has a natural local motion, so u m fo rm lv o ver this quarter and o ver the other. T herefore it is necessary that this
that to every simple natural body, some natural simple motion must be be caused by a resting heaven, one part o f w hich - the one that is o ver us - has
assigned. O n this basis, the empyrean heaven could not be immobile. Sec­ influence and dom inion over the w ell-b ein g o f animals and plants, and the other
ond, the empyrean heaven ought to be nobler than the heavens that are has m ore dom in ion over the gathering ot waters.
moved, because the former is above them, contains them, and confers
powers on them. But it cannot be nobler, because motion is nobler and In rejecting this favorable argument, Buridan suggests that the habitability
prior to rest. Therefore bodies that are moved naturally are nobler than o f the earth might have been ordained from all eternity by God and be in
bodies at rest, which is why the immobile earth is the most ignoble of no way dependent on the empyrean sphere. Indeed, in the very next ques­
bodies. We are thus confronted with a contradiction: the empyrean heaven tion, he explains how the earth’s overall topography could have been pre­
would be both nobler and not nobler than the celestial spheres that are in served eternallv and does so without alluding to the empyrean sphere.''
motion. Finally, although the empyrean heaven is immobile, it could be And in a later question o f the same treatise, when he considers the possibility
moved by God. Therefore it has the potentiality for motion. But if it remains o f the earth's axial rotation, Buridan again rejects the empyrean sphere. To
forever immobile, it would fail to realize that potentiality and would be determine if the earth could rotate axially, Buridan asks whether rest or
perpetually frustrated. "T h u s,” the argument concludes, "it would be ab­ motion is nobler. Rest is nobler, he argues, when a body comes to rest
surd that the heaven should never be m oved.” |S following upon a motion toward its natural place. That is, the rest acquired
Despite these Aristotelian arguments, and "because o f the statements o f after the motion o f a body toward its natural place is superior to the motion
the theologians” (propter dicta theologorum), Buridan found it necessary that brought it to its natural place. B y contrast, motion is nobler than rest
to defend the existence o f the empyrean sphere. After all. Aristotle assumed for those bodies that are always in their natural places and which have no
many things contrary to the Catholic faith, because he sought only to derive other goal than to move in their natural places with their natural motions.
arguments from sensation and experience. Therefore one need not believe
19. “ Et potest responderi ad rationes Aristotelis. quod ipse multa posuit contra veritatem
catholicam, quia nihil voluit ponere nisi posset deduci ex rationibus ortum habentibus ex
17. “ Utrum sit ponendum caelum quiescens sive non motum supra caelos motos.” Bundan sensatis et expertis; ideo non oportet in nmitis credere Aristoteli, scilicet ubi dissonat
[De caelo, bk. a, qu. 6], 1042. 149. The empyrean heaven was a customary topic in sacrae scripturae." Ibid., 152.
commentaries on the second book o f Peter Lombard’s Sememes but also turns up in 20. Ibid., 153.
questions on De eaelo and in commentaries on the Sphere o f Sacrobosco. 21. Ibid., 150-151.
iS. For the three arguments, see Buridar ibid.. 151-152. 22. Question 7, “ Whether the whole earth is habitable." Ibid.. 154-160.
376 THE CELESTIAL REGIO N THE IM M OBILE ORB! THE E M P Y R E A N 377

This, o f course, describes the behavior o f celestial bodies, each o f which exegesis, and authority.28 Thomas o f Strasbourg argued that the existence
revolves with uniform circular motion in its natural place. Since motion is o f the empyrean heaven, which he identified with the tenth orb, could not
nobler than rest for celestial orbs, it would follow by implication (Buridan be shown necessary by reason29 but only by a probable argument.30 Cam -
does not explicitly mention the empyrean sphere) that an immobile em­ panus o f Novara spoke for many when, in his Theorica planetarum, he de­
pyrean heaven would be less noble than the mobile celestial spheres that it clared that “ whether there is anything, such as another sphere, beyond the
surrounds and contains.23 Because no one who accepted the empyrean sphere convex surface o f this [ninth] sphere, we cannot know by the compulsion
would have relegated it to a status more ignoble than that o f the mobile o f rational argument. However, we are informed by faith, and in agreement
celestial spheres contained concentrically within it, we may plausibly con­ with the holy teachers o f the church we reverently confess that beyond it
clude that Buridan’s support for the empyrean sphere was at best dubious.24 is the empyrean heaven in which is the dwelling place ot good spirits. 31
If Buridan’s opposition to the empyrean sphere was ambivalent, Albert Indeed, for some the empyrean heaven had no scriptural basis whatever.32
o f Saxony’s was not. Without naming the empyrean sphere, Albert flatly Its widespread acceptance was based on theological authorities who had
rejects the existence o f an immobile heaven.252 7His three arguments against
6 conceived it as a place for spiritual beings and who were then gradually
it are the same as those reported by Buridan. But whereas Buridan accepted committed to consider further the nature and properties o f that supreme
the immobile empyrean sphere by reason o f faith2" while simultaneously heaven, arriving at those that we have been examining.
furnishing reasons for rejecting it on natural grounds, Albert o f Saxony Although some adopted the empyrean heaven solely by reason o f faith,
ignores theology and,' on the basis o f the three Aristotelian arguments, arguments based on natural philosophy were also formulated for its exis­
rejects the existence o f an immobile sphere, thus becoming one o f the few tence. Without mentioning it by name, preferring to characterize it only as
who did. an immobile orb, Pierre d’Ailly, a staunch supporter o f the empyrean
Was the existence o f the empyrean heaven a matter o f doctrine and faith? heaven, discussed it at some length in his 14 Questions on the Sphere of
An error condemned by the bishop o f Paris in 1244 suggests that it was. Sacrobosco, where, in the second question, he gives three arguments as to
That error, according to Federici Vescovini. is “ the thesis that the glorified w hy an immobile sphere must exist beyond the ten mobile spheres (149V).
soul and the Blessed Virgin are not in the empyrean heaven with the angels The first opinion conceives ot the empyrean heaven as the place and con­
but in the aqueous, or crystalline, heaven and above the firmament, in fact tainer o f the world. It assumes that a mobile sphere must change place either
in the ninth sphere.’ ' In Paris, at least, the empyrean heaven seems to have as a whole or with respect to its parts, from which it follows that the mobile
had Church sanction. A long tradition o f general acceptance throughout sphere is in a place. The place o f that mobile sphere must surround and
Christendom may, in any event, have conferred upon the empyrean heaven contain it. Moreover, that surrounding place must also be immobile, a
a kind o f quasi-doctrinal status. If Albert o f Saxony violated church tradition consequence which d’Ailly does not actually establish. But the idea that all
or doctrine, there is no evidence that he suffered any penalty or adverse movable things ought to have an ultimate immobile place had earlier mo-
reaction.
28. In his Sentences [bk. 2, disc. 2, qu. 2. art. 1], 1929-1947. 2:71. Thomas Aquinas explains
that “ the empyrean heaven cannot be investigated by reason because we know about the
2. In defense o f an empyrean sphere heavens either by sight or by motion. The empyrean heaven, however, is subject to
neither motion nor sight. . . but is held by authority.
N o one can doubt that scholastics overwhelm ingly accepted the existence 29. As the first o f two conclusions in his Sentences, bk. 2. dist. 2, qu. 2, 1564, 133V, col. 1.
ot an empyrean sphere. And yet, because it was an invisible entity without Thomas o f Strasbourg declares: “ quod caelum decimum esse, quod sancti appellant em-
pyreum, non potest ostendi necessaria ratione.
detectable effects, belief in its existence was based largelv on conjecture, 30. Thomas o f Strasbourg’s second conclusion (ibid., 133V, col. 2) asserts “ quod caelum
empyreum esse potest aliquo modo declarari probabili persuasione.
23. These arguments appear in question 22, “ tertia persuasio” ibid., 228, for the arguments 31. Campanus o f Novara, Theorica planetarum. 1971. 183. I have made one change in the
in favor o f the greater nobility o f rest, and 230. lines 25-31, for the conditions under translation by Benjamin and Toomer. Following “ rational argument” they add “ alone,”
which motion is nobler. For more on the comparative nobility o f motion and rest, see which implies that rational argument, along with other methods, is used to arrive at the
this volume. Chapter 20, Section V.2.b. existence o f the empyrean sphere. In fact. Campanus. like Thomas Aquinas before him.
24. As we shall see in Section IV ot this chapter, Amicus used the rest—motion and noble- was arguing that rational argument was useless in determining the existence o f that special
ignoble dichotomies to defend the existence of an immobile empvrean heaven. sphere.
25. Albert o f Saxony [De caelo. bk. 2, qu. 8], 1518, loyr, col. i-io y v , col. 2. 32. Dictionnaire de theoloyie catholiijue. vol. 2, pt. 2. cols. 2505, 2508. Among those who tound
26. Buridan says that he accepts a resting, or immobile, heaven that lies beyond the mobile no scriptural basis tor the empyrean sphere were Thomas Aquinas. Durandus de Sancto
heavens “ because we assume an empyrean heaven there on faith (Arguitur quod sic, Porciano (ca. 1275-1334), and Cajetan (Thomas de Vio; 1468-1534). For Dante, it had
quia ex fide nos ponamus ibi caelum empvreum). Buridan, De caelo, bk. 2, qu. 6, 1942, no real existence in space but only in the Divine Mind (see Conoioio. II.11i.II). For this
149. reference and for a thorough set o f citations to Dante's treatment o f the empyrean heaven,
27. See Peter o f Abano, Lucidator, 1988, 200. see Toynbee. 1968, 181 (“ Cielo empireo").
37-S THE CELESTIAL REGION THE I M M O B I L E ORB; THE EM P Y R E A N 379

tivated Campanus o f Novara to declare (1971, 183) that “ The empyrean’s pvrean sphere seemed remote and uninvolved with the governance o f the
convex surface has nothing beyond it. For it is the highest o f all bodilv world. Uncertainties about it were also attributable to the fact that “ the
things, and the farthest removed from the common center o f the spheres, saints,” as Saint Bonaventure expressed it, “ speak little o f this heaven be­
namely the center o f the earth; hence it is the common and most general cause it is hidden from our senses, and the philosophers [sav] even less.” 3”
‘place’ for all things which have position, in that it contains everything and From the thirteenth century onward, however, discernible efforts were
is itself contained by nothing.’’3’ Those who assumed this argument - and made to attribute some effects to the empyrean heaven and thus involve it
all who believed in the existence o f an empyrean orb did - agreed with in cosmic operations. Scholastics began to inquire whether the empyrean
Aristotle that our world was surrounded by an ultimate convex surface but heaven could influence inferior things. There was nonetheless some reluc­
departed from him by the assumption o f its im m obility.3 343
*
5 tance to involve the empyrean sphere in regular cosmic operations. Bar­
D ’A illy’s second argument is based on Aristotle’s claim (Qe caelo, bk. 2, tholomew the Englishman, for example, denied that the empyrean heaven
ch. 2) that the heavens possess absolute differences in directions, which he was needed for the continuation of generation among inferior things.37 One
identified as right and left, front and back, and above and below. D ’Ailly o f its functions was, rather, to complete the body o f the universe by serving,
insists that mobile spheres could not exhibit such directions if they turned, along with the earth, as one o f a pair o f immobile, bodily extremes. Whereas
because that direction which is now right would become left and the part the earth was opaque (and presumably the heaviest body in the universe),
that is up would become down. O nly in an immobile sphere can such the empyrean was the most luminous, most subtle, and least heavy body.
absolute directions and differences o f position be found. In his commentary on Sacrobosco’s Sphere, Michael Scot describes the em­
The third argument concerns possible celestial influences. Could an im­ pyrean sphere as immobile and uniformly filled with light, a light that can
mobile empyrean heaven influence terrestrial change and perhaps even be extend its influence only to the inferior heavens, though not uniformly,
essential tor that purpose? If so, it would not only have a theological role because those less perfect heavens are incapable o f receiving the empyrean
but an important and vital cosmic role as well. Because o f the potential light in a perfect w a y .38
signiticance o f such an empyrean heaven, we shall consider this problem, N ot all were so cautious or negative. Indeed, the anonymous thirteenth-
including d 'A illv’s third argument, at some length. century author o f the Suttima philosophiae, falsely attributed to Robert Gros­
seteste, made the empyrean sphere the most noble and powerful body in
the universe. “ The empyrean heaven,” he declared, “ is the original principle
o f rest o f all natural things.” Moreover, it was the empyrean heaven, not
III. C an the em p yrean h eaven influence the center o f the earth, with respect to which all the mobile spheres originally
the terrestrial region? moved. O n this basis, he proclaims that "the immobility o f the empyrean
Until the seventeenth century, when some raised doubts about its exis­ heaven is more the universal cause ot every transmutation of generable and
tence,3’ controversy about the empyrean sphere focused not on its existence corruptible things than the primum mobile and the other inferior spheres,
but on its function and influence. Although in a subsequent chapter (Ch. just as a primary cause is more a cause than a secondary cause.” 3'3
19), we shall address the crucial theme o f celestial influence as it pertained Although most scholastics assigned a causal role to the empyrean heaven.
to all the mobile orbs, in this chapter we shall consider the subject o f celestial
influences with reference only to the immobile, empyrean heaven. 36. “ Dicendum. quod quamvis Sancti parurn loquantur de hoc caelo. quia latet nostros sensus.
et philosophi adhuc minus.” Bonaventure [Sentences, bk. 2. dist. 2, art. 1, qu. 1J, Oy ra,
Throughout the five centuries embraced by this study, ambivalence best [885, 2:71, col. 2.
characterizes scholastic attitudes about alleged influences o f the empyrean 37. Bartholomew the Englishman. De rerum proprietatibus. bk. 8. ch. 4. 1601, 380.
heaven. As an invisible, immobile entity whose essential raison d’etre was 38. “ Et primum celum a theologis dicitur empvreum non ab ardore sed a splendore et est
umformiter plenum lumme et immobile, non uniformiter tamen induit lumen suum in
purely theological and which lacked explicit biblical justification, the em- inferioribus celis eo quod actio agentis non recipitur in passum per modum ipsius agentis
sed etiam per modum patientis. ut dicitur in libro De substantia orbis.” Michael Scot [Sphere,
33. This passage is also translated in Duhem [Anew], 19S5, 17s. Dante also savs that the lec. 4], 1949. 283.
empyrean heaven “ contains all bodies and is contained by none.” See Tovnbee, 1968. 39. In chapter 3, tractatus 15, the anonymous author o f the Summa philosophiae treats o f the
181, for references. In Chapters S and 9, we saw the kinds o f things that were conceived empvrean sphere (see Grosseteste, Summa philosophiae, 1912, 545-548). The Latin texts
to exist, or possibly exist, beyond the world or its last convex surface. o f the two translations just cited are respectively: "Caelumque empvreum totius quietis
34. On Aristotle’s discussion o f the place o f the last sphere, see Chapter n. Section III. Duhem rerum naturalium originate esse principium” (ibid., 546) and “ Quietem caeli empyrei
considers the empyrean heaven within a general treatment o f the medieval doctrine or magis esse causam universaliter omnis transmutationis generabilium et corruptibilium.
place Lc Systeme, 1913-59. 197-202. quam primum mobile ceteraeque sphaerae inferiores, stcut causa prim a magis est causa
35. For example, Thomas White, who rejected it, and Nicolaus Caussin, who questioned its quam causae secundariae" (ibid.. 547). For a French translation o f tf second passage
existence. Donahue, 1972, 253, 251. (and considerably more), see Duhem, Le Systeme, 1913-1959, 7:199—2.0.
380 THE CELESTIAL REGION THE IM M O BILE ORB! THE EM P Y R E A N 381
few were prepared to go so far as to make it “ the universal cause o f every movable sphere, the empyrean sphere is also the indirect cause o f the gen­
transmutation.” In the question, “ Whether the empyrean heaven influences eration and corruption caused by all the celestial spheres below the first
inferior things” (Utrum caelum empyreum influat in haec inferiora), Saint movable sphere.
Bonaventure presents five arguments in favor o f influence and five against.40 Like Bonaventure and Thomas Aquinas, most theologians expressed
He concludes that “ any o f these opinions is sufficiently probable. But which opinions on the empyrean heaven. Without providing much detail, Richard
is more true is not clearly apparent.” 41 Neither reason nor authority can o f Middleton ([Sentences, bk. 2, dist. 2, art. 3, qu. 3], 1591, 2:44-4$), who
make obvious w hy the empyrean heaven has to exercise any influence on unhesitatingly accepted the influence o f the empyrean on inferior things,
sublunar things, “ since the influence o f the other inferior [celestial] bodies suggests that the immobile empyrean sphere moderates the influences ex­
seems to suffice.” Nevertheless, Bonaventure inclined toward acceptance erted by the mobile celestial spheres. The empyrean sphere was ordained
o f an empyrean influence on inferior things, justifying his decision by ob­ for the use o f man both in the future life, when it would presumably serve
serving that among all bodies, the empyrean heaven was the first created as the abode o f the elect, and in the present life, when it influences inferior
and has the greatest size (moles) and power (uirtus). Its size enables it to things by its power. Also convinced o f an empyrean influence on inferior
locate all bodies by surrounding and containing them; its power, or influ­ things, Thomas o f Strasbourg sought to convince his readers o f its necessity,
ence, enables it to animate and conserve things, although it may achieve although he fails to describe the form o f that influence.45 Recognizing that
this through the other celestial spheres.4' the chief obstacle to the concept o f influence stemming from an immobile
Despite a similar initial ambivalence, Thomas Aquinas eventually arrived empyrean heaven was the widely accepted idea that only a body in motion
at the same conclusion. Because he accepted Aristotle’s principle that only can influence another body,46 Thomas counters by invoking the Joshua
bodies in motion could affect other bodies, Thomas at first (in his com­ miracle, which, as we shall see, was frequently cited as evidence that even
mentary on the Sentences) denies that an immobile body like the empyrean when at rest the celestial region could influence inferior things. Although
sphere could influence other bodies.43 But later, in his Quodlibetal Questions, the Sun had been commanded to halt for a day, celestial influences continued
he concedes that, upon further reflection, the empyrean heaven does indeed to affect the earth, “ for otherwise, those living here below would have been
influence interior bodies.44 The world order demands that corporeal things dead.” 47 Similarly, a magnet at rest attracts iron.48 Thus a body at rest is
be governed by spiritual things and inferior bodies be ruled by superior capable o f exerting influence on distant things. Thomas o f Strasbourg there­
bodies. It would be absurd if the empyrean heaven did not influence inferior fore insists that not only can the immobile empyrean heaven influence
bodies, for then it would form no part o f the universe. inferior things but that it can do so directly, without the mediation o f
T o explain how the empyrean sphere influences inferior bodies, Thomas celestial motions. That this does not actually happen is a consequence o f
employs a hierarchy o f descending perfections based on rest as most perfect, the nature o f inferior things, which can only receive empyrean influences
followed by uniform motion, and finally difformity — that is, nonuniformity indirectly through the mediation o f the celestial motions.
- o f motion. B y its rest, the empyrean heaven causes the first movable D ’A illy ’s third argument, alluded to earlier, seemed to offer empirical
sphere to move with a uniform motion and thus produce the daily motion evidence in favor o f an immobile heaven. In its various versions, it was
as it moves from east to west; the first movable sphere, in turn, causes all frequently repeated in the sixteenth and seventeenth centuries. Citing dif­
the spheres below it to move with motions that vary with the distances o f ferences in fruits and customs and in many other things that one finds on
their planets from the earth. As the planets vary their distances, they cause parts o f the earth that lie on the same latitude, or, as d’Aillv expressed it,
generation and corruption in sublunar bodies. Thus the empyrean sphere
is the direct cause o f uniform motion and therefore the direct cause ot 45. Thomas ot Strasbourg, Sentences, bk. 2, dist. 14, qu. 1: “ Whether the heavens are the
permanence in the universe, and through the uniform motion o f the first cause ot'inferior things” (“ An caelum est causa horum inferiorunt” ), 1564, 156V, col. 2-
158V, col. 2; for the discussion o f the empyrean heaven, see I58r, col. 2 -1 58V, col. 2.
40. Sentences, bk. 2, dist. 2, art. 1, qu. 2, Opera, 1885, 2:73-75. 46. As Thomas o f Strasbourg explains (ibid., I58r, col. 2): “ quidam venerabilis doctor in
41. “ Quaelibet haruin opinionum satis probabilis est; quae autem sit magis vera, non plane suo scripto dist. 2 secundi libri Sententiarum quia eo ipso quod celum empyreum non
apparet.” Ibid., 74, col. 2. movetur non potest sibi competere aliqua realis influentia in cetera corpora quia corpus
42. Ibid., 74, col. 2-75, col. 1. non agit nisi per motum.”
43. “ Respondeo dicendum quod caelum empyreum nullam habet influentiam super alia cor­ 47. “ Ad primum dicendum quod nec ilia consequentia valet nec eius probacio quia tempore
pora. quae rationabiliter poni possit.” Thomas Aquinas. Sentences, bk. 2, dist. 2, qu. 2, Iosue sol non movebatur per spacium unius diei et tamen habuit realem influentiam in
art. 3, 1929-1947, 2:76. ista inferiora alias ista viventia hie inferius fuissent mortua.” Ibid., 158V. col. 1. Among
44. Quodlibetum 6, qu. 11, art. 19: “ Whether the empyrean heaven exercises influence over those who cited the Joshua miracle as evidence that the immobile heavens could influence
other bodies.” See Thomas Aquinas, Quaestiones quodlibetales, 1949, 130; also Litt, 1963- inferior things were Hervaeus Natalis in the thirteenth century, Nicole Oresme in the
260-261. Thomas wrote his Sentences during 1252-1256 and his sixth quodlibetal question fourteenth, and the Coimbra Jesuits in the sixteenth.
in 1272 (see Weisheipl, 1974, 358-359, 367). 48. “ Magnes non motus trahit ferrum." Ibid.
3-S'2 THE CE LEST IA L REGION THE IM MOBILE ORB: THE E M P Y R E A N 3^3

that lie “ between east and west [equidistant from the poles,” d’A illy insists thoughts in mind that Otto von Guericke cited Jacques du Bois, an eccle­
that these differences cannot be explained by the circular motions o f mobile siastic o f Leyden, as insisting that anyone “ who denies that there is an
spheres, because, as the same configurations o f celestial bodies swept over empyrean [heaven] above the visible heavens does not believe the sacred
the same latitude, all the inhabitants and all the fruits and vegetation along words nor does he believe the foundations o f Christianity” (Experimenta
that latitude would be exposed to the same celestial influences. Under these nova, 1672, 49). '4
circumstances, samenesses rather than differences ought to be found among The empyrean was also a perennial problem for Aristotelians because it
members of a given species on the same latitude. But an immobile sphere confronted them with the problem o f whether motion or rest was nobler.
like the empyrean could produce the differences we see. It could act dif­ Was an immobile celestial body, even one that surrounded all o f the moving
ferentially on various parts o f the earth and, by virtue o f its immobility, spheres, more noble than the moving spheres it contained? The kind ot
do so constantly and permanently.v> As evidence o f such differential power, argument that Buridan formulated found its counterparts in early modern
d’Ailly repeats a common belief that all stars have greater power in the east scholasticism. Amicus ([De caelo, tract. 4, qu. 6, dubit. 5], 1626, 194, col.
than m any other part o f the heaven. '0 1), for example, mentions an objection to the existence o f the empyrean
heaven, namely that mobile bodies are nobler than immobile bodies, as is
obvious from the earth, which is immobile and deemed the most ignoble
o f all bodies. If the empyrean heaven is said to be the noblest o f all spheres,
IV . C oncepts o f the em p yrean o rb in the late sixteen th and it cannot, therefore, be immobile. Amicus resolves the dilemma (ibid., 197,
the seven teenth cen tu ry col. 2) by emphasizing the virtue o f immutability. The empyrean heaven
The range of ideas about the empyrean heaven that were predominant in is not immobile because it lacks the capacity for motion but is immobile
the late Middle Ages remained prevalent in the late sixteenth and the sev­ bv divine decision, since it does not require motion to realize its own ends.
enteenth century, although some new interpretations and attributes were Amicus further denies that an immobile body in its proper place is more
added. Because it tailed to manifest its existence in any o f the usual ways, ignoble than a mobile body in its proper place. Indeed, immobility is a
there was a continuing sense that the existence o f the empyrean heaven was perfection, as is obvious from the fact that God is immutable and therefore
incapable of rational p ro o f.'1 Although a rigorous demonstration was un­ immobile. But motion is an imperfect thing, because it is a path to perfection
attainable, certain kinds of evidence served in its stead. In one o f the most and not the perfection itself. The peculiar demands o f the empyrean heaven
thorough and detailed treatments o f the empyrean heaven, Bartholomew fostered an ambivalent attitude toward the contraries rest and motion.
Amicus presents the kinds o f reasons that were deemed persuasive. Bv the Whereas Buridan exalted motion over rest, which he compared to the ig­
seventeenth century. Amicus could appeal for acceptance o f the empyrean noble earth, Amicus linked rest with divine immutability and exalted im­
heaven to the authority of the Church Fathers, the traditional acceptance mobility over motion."'
of the scholastics, and the common acceptance o f the Church. '2 With such The invisibility o f the empyrean heaven was also puzzling. If it really
weighty credentials, few if any scholastic authors were inclined to deny the existed, w hy was it not visible, since it is said to be the moat lucid body?
existence of this immobile orb. Indeed. Amicus also assumed, along with Most responses emphasized the great distance o f the empyrean heaven and
many others, that when Genesis declares that God created heaven and earth, the thickness o f the inferior heavens that were interposed between it and
this signified either the empyrean heaven only or the latter along with all us. For such reasons, the brilliant splendor o f the empyrean was impercep­
the other heavens. Moreover a number o f biblical passages speak o f this tible to human observers. Pedro Hurtado de Mendoza appealed to the
heaven as the domicile o f God and the blessed, and the “ heaven" in question divine when he suggested that God had made a certain opaque curtain to
was usually assumed to be the em pyrean." It was probably with such conceal the empvrean heaven from unworthy eyes in a manner similar to
the silken cover used to conceal the Holy ot H olies.'
49. We saw. in Section II. 1 ot this chapter, that Buridan rejected this argument.
50. Mastrius and Bellutus [De coelo, disp. 2, qu. 7, art. 1 1, 1727, 3:310, col. 2, par. 1X9-311. 54. Also in the seventeenth century. George de Rhodes insisted that the empyrean heavci
col. 1. par. 190. repeat this in the seventeenth centurv. must not be denied bv any Catholic. De Rhodes. De coelo, bk. 2. disp. 2. qu. i, sec. 2,
51. See Donahue's translation o f a relevant passage (1972. 233) from p. >07 o f the De coeio pt. 3, 1671. 2X1, col. 1.
section of Roderigo de Arriaga's Cursns philosophicns. 33. Nicole Oresme also considered rest nobler than motion (see Oresme [De spero, qu. S|.
>2. “ Prima conch: Certum dari coeium empvreum. Pnmo patet ex auctoritate Patrum et 1960a. lOi). Scholastic authors sometimes assigned greater nobility to rest or motion
Scholasticorum et commum assensu Ecclesiae.” Amicus [De eoelo. tract. 4. qu. 6, dubit. solelv on the basis o f the argument they sought to demonstrate.
3, art. 2), 1626. 194, col. 1. This is the first o f six conclusions that represent Amicus's 50. See Amicus. De caelo, tract. 4, qu. 6, dubit. 3, art. 4, 1620, 19". col. 2.
own opinions and extend over pages 193, col. 2—197, col. 2. 57. Hurtado de Mendoza f De coelo, disp. 2. sec. sj. 1013, 375, col. 1. par. 63. In his discussion
33. Amicus cites Psalms 67. 102. 113. and Deuteronomy 10. o f the empyrean heaven, von Guericke, Lxperimento novo. 11172. 49, cols. 1—2. repeats
3«4 THE CELESTIA L REGIO N THE I M M O B I L E ORB! THE EM P Y R E A N 385
The central issue about the empyrean heaven was not, however, its ex­ late sixteenth and the seventeenth century the response was divided: some
istence, which almost all scholastics assumed, but rather its alleged effects denied causal efficacy to the empyrean, while others upheld it. Let us ex­
on inferior parts o f the universe, ranging from the orb immediately below amine the latter position first.
it all the way to the earth itself. Late sixteenth- and seventeenth-century
scholastics adoped a variety o f opinions, ranging from a denial o f causal
efficacy - Raphael Aversa, Mastrius and Bellutus, and Illuminatus Oddus a. The affirmative arguments
- to the acceptance o f direct influence on terrestrial events - Christopher For those who believed in the causal efficacy o f the empyrean heaven, the
Clavius, the Coimbra Jesuits, and Bartholomew Amicus. O n balance, how­ problem was to identify the direct effects that it caused. Because its sublunar
ever, there was a shift o f opinion toward a denial o f terrestrial influence to effects were hidden, the Conimbricenses, for example, professed ignorance
the immobile empyrean heaven.
as to the manner in which the empyrean heaven acted on inferior things.'10
Others, however, were more forthcoming. In support o f an active empyrean
i. Does the empyrean heaven cause terrestrial effects? heaven that helped govern the sublunar region, some, like Bartholomew
Amicus, invoked proof “ by natural reasons” (rationihus naturalibus), a kind
In a manner similar to Thomas Aquinas, the Coim bra Jesuits, for example, o f proof that had by then become rather common and supplemented tra­
argued that if the empyrean heaven were to be considered part o f the uni­ ditional appeals to theology and Scripture. Such “ proofs” involved the
verse, as they believecf, influences o f some kind had to flow from it, for identification o f discernible terrestrial, or sublunar, effects that were attrib­
otherwise it would not be part o f the world, and, according to Mastrius utable to the empyrean sphere and could thus reinforce belief in its real
and Bellutus and others, its existence would be in vain. Believers in an existence. The most dramatic evidence lay in the diversity o f effects observed
active empyrean heaven also argued, as Raphael Aversa informs us ([De in different parts o f the earth and even on the same parallel o f latitude. For
caelo, qu. 35, sec. 5], 1627, 180, col. 1), that because the substance o f the instance, men o f a given region have the same inclinations, whereas those
empyrean heaven is nobler than that o f all other orbs below it, it must o f another have different inclinations; and the trees o f one region are alike,
surely influence terrestrial change and events just as they do. Otherwise it whereas those o f another are different. In common examples, it was claimed
would be more ignoble, rather than more noble and efficacious, than the that although the fastest horses are bred in Hungary, on the forty-seventh
orbs below it. T o counter this argument and deny causal efficacy to the parallel, elsewhere on the same parallel they are not produced at all; apes
empyrean heaven, Mastrius and Bellutus argue that the empyrean sphere’s (.simiae) are generated in Mauretania but not elsewhere on the same parallel.
perfect substance does not imply a more perfect operation,sX since a perfect Such diversity along the same parallels ought to be ascribed to the heavens,
substance need not exercise influence on anything below it. Despite a lack since it appears unlikely that they arose because o f differences in the earth
o f causal efficacy on inferior things, the empyrean sphere would not have itself. But are the heavens that can explain these diversities mobile or im­
been created in vain, because it serves as the abode o f the blessed and mobile? M obility cannot explain them, because mobile heavens sweeping
produces a light appropriate to that region.59 uniformly and regularly over the same parallel would produce uniform,
But what kind o f influences, if any, could the empyrean heaven exert on rather than diverse, terrestrial effects in every part o f the earth under that
the regions below, especially the earth? We saw earlier how Pierre d’Ailly parallel. The cause o f these differences ought therefore to be attributed to
responded to this fundamental problem, and w e shall now see that in the the empyrean sphere, which, because o f its immobility, could focus rays
from every one o f its parts to the corresponding region o f earth beneath. "
this opinion when he declares: “ Some state that the farthest part o f the empyrean is solid T o produce diverse effects, however, different powers had to be assumed
and is darkened from [or bv?[ the inferior [or nearest] part as if by a certain curtain of
to radiate from different parts o f the empyrean heaven. But if, as most
thick and opaque matter, lest that light and [that) heaven be perceived [or seen) by
unworthy mortals here in the world; just as once among the Jews, the Holy o f Holies
was concealed with a certain cover." 60. See Conimbricenses [De caelo, bk. 2, ch. 3, qu. 2, art. 2], 1598. 196.
58. Mastrius and Bellutus. Decode, disp. 2, qu. 7, art. 1, 1727, 3:510, col. 2, par. 189-511, 61. The substance o f this argument with a variety o f examples appears in Amicus, De caelo,
col. 1, par. 190. tract. 4, qu. 6, dubit. 5, art. 2, 1626, 195, cols. 1-2 and tract. 6, qu. 3, 353, cols. 1-2;
59. Illuminatus Oddus [Dc coelo, disp. 1, dub. 24], 1672, 74, col. 1, accepts the same argument Clavius [Sphere, ch. 1], Opera, 1611, 3:24; Aversa. De caelo, qu. 35, sec. 5, 1627, 180,
- indeed he even mentions Mastrius and Bellutus (71, col. 2) - when he declares that col. 2, who mentions Clavius as his source for the argument; and Mastrius and Bellutus,
“ perfection in a substance does not always imply a more perfect operation. And this is De caelo, disp. 2, qu. 7, art. 1, 1727, 3:510. col. 2, par. 189. Clavius, Mastrius and Bellutus,
obvious in angels, who, although they are most perfect, can elicit no substantial action. and Illuminatus Oddus, De coelo, disp. 1. dub. 24, 1672. 73, col. 2, mention the examples
Thus it suffices that it [the empyrean heaven] should serve as a dwelling place for the o f Hungary and Mauretania. Although the same examples are sometimes described dif­
blessed so that it cannot be said to be in vain." Aversa says much the same thing (De ferently. their object is to show that only the immobile empyrean heaven could be the
caelo, qu. 35, sec. 5, 1627, 180, cols. 1-2). cause o f such diversity.
3 S6 THE CELESTIAL REGION THE IM M OBILE ORB: THE EM PYREA N 387

believed, the empyrean heaven is homogeneous, its impact should be the The argument that dramatically characterized the empyrean heaven as
same all over. How, then, could such a homogeneous body radiate different heterogeneous so that its influences could explain the diverse customs,
influences from its different parts? Recognizing a potentially troublesome plants, and animals that lay along a given parallel o f latitude also came under
problem, Amicus, who defended the causal efficacy o f the empyrean heaven, attack. The idea o f describing the empyrean heaven as heterogeneous must
denied true homogeneity to it, arguing that “just as diversity in inferior have struck Raphael Aversa as contrary to one’s expectations about the most
things belongs to the perfection and pleasure o f sight, so in the empyrean perfect ot all celestial spheres, which he regarded as the same in all its parts."16
heaven,” that is, just as the perfection o f sight is partially constituted from Rather than ascribe heterogeneity to the empyrean heaven, Aversa assigned
the diversity o f things seen, so also is the perfection ot the empyrean sphere the cause o f terrestrial diversity along a given parallel o f latitude to the earth
partially constituted from the diversity o f inferior things that it helps pro­ itself, which possessed varied dispositions, different seeds o f things, and a
duce by its unhomogeneous nature. Amicus was not alone in attributing multiplicity o f mixtures o f water and air within the bowels o f the earth.
heterogeneity to the empyrean heaven/1' These various dispositions influence events on the earth’s surface and can
cause diversity/1' Indeed, even celestial bodies moving uniformly over the
b. The negative arguments earth’s surface might cause ditterential effects. Oddus emphasizes the dis­
parate relationships that distinct parts o f the earth may have with celestial
The sense that the empyrean heaven was not created for the purpose o f bodies. Conjunctions from the celestial region may affect different terrestrial
influencing sublunar things remained strong. Those who adopted this at­ areas differentially. “ Ifefficient and material causes vary,” Oddus concludes,
titude had to reply to the argument that if the empyrean heaven exercised “ it is little wonder that some effects are generated in one area and not in
no influence on the physical things it contained, then the most perfect another. ” 6X
substance God had created - and it was routinely so described — was created Some assumed that if the empyrean heaven affects the sublunar region,
in vain. This charge was easily countered by insisting that G od’s primary it does so by somehow transmitting effects to the other celestial spheres
purpose in the creation ot the empyrean heaven was to provide a domicile below. But Aversa wonders what it is that the empyrean transmits to those
for the bodies and spirits o f the blessed. Its function was supernatural, rather inferior celestial orbs. After all, the latter possess their own proper qualities,
than natural/1’ And o f course it was always thought to function as the attributes, and causal powers and are not in need o f anvthing from a su­
container o f the world/’4 But even as the most perfect o f things, it was not perior, immobile sphere. In sum, the empyrean has nothing to confer on
essential that it engage in more perfect operations, such as influencing ter­ the mobile spheres below.
restrial things more pertectly than did the mobile celestial bodies it con­ If the transmission o f physical influences by the empyrean heaven was
tained. To illustrate the point, Illuminatus Oddus noted that although angels highly controversial, the transmission ot light was more readily conceded,
are the most perfect o f things, they do not cause substantial actions.'”1 because o f the alleged brilliance and splendor o f the empyrean heaven. But
62. Thomas Compton-Carleton \De coelo, disp. 2, sec. 4], 1649, 404, col. 2, denied homo­
this particular influence is discussed later, in the chapters on celestial light
geneity to the empyrean sphere when he declared: "it seems to me that a variety o f effects (Ch. 16) and the nature o f celestial ettects on the terrestrial region (Ch. 19),
that are perceived in different parts o f the earth are poured forth from the diverse, where light is distinguished as one ot three basic modes o f influence.
heterogeneous parts o f the empyrean heaven.” Melchior Cornaeus [ D p coelo, tract. 4,
disp. 2, qu. 1. dub. 5], 1657, 491, also seems to have considered the empyrean heaven
heterogeneous (quasi heteroqeneas) when he divided it into three parts, the lowest o f which
is solid and serves as a foundation for blessed bodies: the middle is fluid and respirable, 2. The status o f the empyrean heaven
like air; the highest part is also solid and encloses the respirable, airlike substance and
serves as the roof o f the celestial structure. As the source o f this threefold division o f the From the thirteenth to the seventeenth century, few scholastic theologians
empyrean sphere, Cornaeus cites Lessius’s De sinmno lunio, ch. 8. denied the existence o f the empyrean heaven, and several, as we saw, even
63. Aversa, De carlo, qu. 35, sec. s. 1627, 180, col. i, declares that "iam probatur non esse
de facto ullam operationem tribuendam celo empireo in haec inferiora corpora quia caelum
empireum per se et proprie conditum et ordinatum fuit ut esset domicilium Beatorum.” 66. “ Nam sine dubio hoc caelum [i.e., the empvrean] est totum homogeneum et eiusdem
64. Oddus, De coelo. disp. I, dub. 24, 1672, 73, cols. 1-2, explains that although the empyrean rationis in omnibus suis partibus.” Aversa, De caelo. qu. 35, sec. 5, 1627, 1S1, col. 1.
heaven lacks the capacity for action, it serves to contain the bodies ot the blessed and the This argument and the others in this and the next paragraph are drawn from the sixth
blessed spirits and is the boundary for the physical universe (“ Empireum quanavis omnis o f seven arguments Aversa musters to deny claims for an active empyrean heaven. For
actio ab eo removeatur, sed satis est deservire ad continendum intra se corpora beatorum all seven arguments, see ibid., 179, col. 1-1S1, col. 1; the sixth appears on 180. col. 2—
et domicilium esse beatorum spirituum, et machinam hanc umversi terminare” ). 181, col. 1.
65. "Nam ut saepe dictum est, perfectio in substantia non semper arguit perfectiorcm op­ 67. Oddus, De coelo, disp. 1, dub. 24, i 6~2. 73, col. 2. says much the same thing. The earth's
erationem. Et patet in Angeiis, qui licet perfectissimi sint, nullam tamen possunt actionem powers and dispositions vary trom region to region and thus may plausibly explain
substantialem eiicere. Unde sutficit ut beatis ad inhabitandum inserviat, ne otiosum di- different effects in different regions.
catur." Ibid., 74, col. 1. 68. Ibid.
3*8 T H E C E L E S T IA L R E G IO N THE IM M O B ILE ORB! THE E M P Y R E A N 389

went so far as to insist that rejection o f the empyrean heaven was tantamount o f the Copernican revolution. Absent are the names o f Copernicus, Tycho,
to a rejection o f Christianity. All accepted its theological function as the Galileo, Kepler, Newton, and all the others. Although his body was present
abode o f the blessed, but disagreement arose when it came to assigning a in the eighteenth century, H idalgo’s mind, filled with the thoughts o f Ae­
cosmological role to the empyrean sphere. It was one thing to concede its gidius, lived only in the thirteenth century. Perhaps there were a few more
existence, and quite another to believe that it influenced the physical world. like Hidalgo in the eighteenth century, who defended the empyrean heaven
The absence o f discernible effects, its acknowledged invisibility, and its as an active influence in the physical world. Already in the seventeenth
patently theological nature made arguments about the cosmological role o f century, however, most o f their scholastic predecessors had eliminated it
the empyrean heaven inconclusive. Whereas thirteenth-century theologians from their cosmology.
and natural philosophers were prepared to attribute some influence on the
physical world to the empyrean heaven, many denied it that capability by
the seventeenth century. In the numerous commentaries or questiones on De
caelo written by scholastic theologians in the seventeenth century, either as
independent treatises or as part o f a larger cursus philosophieus, many - like
Sigismundus Serbellonus, Francisco de O viedo, and Johannes Poncius -
omitted serious consideration o f the empyrean heaven/*9 For them it may
no longer have seemed appropriate to include such a blatantly theological
heaven in a cosmological treatise.
As we saw, however, serious discussions continued. A m ong those who
still thought it worthy o f discussion in a cosmological treatise, as many
seemed to reject its influence as proclaimed it. When account is taken o f
those who ignored the subject and those who opposed any empyrean in­
fluences, we may conclude that the number o f scholastic theologians who
believed that the empyrean heaven influenced celestial and/or terrestrial
physical operations was relatively small by the middle o f the seventeenth
century. Copernican cosmology may have played a significant role in pro­
ducing this result. But not even Copernican cosm ology and the Newtonian
Scientific Revolution could completely cause the disappearance o f defenders
o f an empyrean influence. Well into the eighteenth century, Juan Hidalgo,
a Spanish Augustinian Hermit, who wrote a cursus philosophicus “ according
to the thought o f Blessed Aegidius Rom anus,” vigorously repeated his
master’s thirteenth-century defense o f empyrean physical influence (1737,
2:73, col. 2-76, col. 1). There are no surprises in H idalgo’s discussion, much
o f which is directed specifically against Aversa. As the most noble o f all
bodies, the empyrean heaven should influence inferior things, not by contact
but by its operation. Although the empyrean heaven is immobile, its influ­
ence is disseminated by the motions o f the inferior celestial spheres. As a
part - indeed, the noblest part — o f the universe, the empyrean heaven must
exercise an influence on other parts. The philosophers have mistakenly
accounted for the behavior o f inferior things by assigning all the causes to
the mobile heavens. But they should attribute some influences and causal
changes to the empyrean sphere, which was discovered by revelation and
ought to be properly fitted into the cosmic picture. H idalgo’s readers would
find no mention o f the great figures who were responsible for the triumph

6y. Others, like Pedro Hurtado de Mendoza, treated it only briefly.


C E L E S T IA L L IG H T 391
1 6 body; whether light as the quality o f a body is a substantial or accidental
form; whether the light that emanates trom a body is itself a body; and
whether the light that emanates from a body is a substantial or accidental

Celestial light form .' Bonaventure’s responses reveal no particular concern for the rela­
tionship o f light to celestial bodies.2 Thomas Aquinas and others posed
similar questions in their commentaries.3 Although the responses to these
questions occasionally contained material relevant for cosm ology,4 most
were concerned with the nature ot pure light, with little regard for the role
o f celestial bodies.
As a theme and metaphor, light has been perhaps the most pervasive and Comm ents on the fourth day ot creation produced similar results. Here
ubiquitous topic in Western science, philosophy, theology, literature, po­ the questions are about the “ luminaries,” or visible heavenly bodies created
etry, and art. Its chiet significance for cosm ology, however, lies in its on the fourth day. The comments are usually about their planetary natures,
relationship to celestial bodies. their motions, and how they compare to terrestrial bodies, but rarely about
their relationship to light.3 Certain questions about the empyrean heaven
concerned light. One that was usually posed, in commenting on the second
distinction o f the second book o f the Sentences, asked “ Whether the em­
I. T h e sources
pyrean heaven is lum inous.” Although there was general agreement that
Light as it was treated in the tradition o f geometric optics and in visual the immobile, invisible, empyrean heaven was the location o f purest light,
theory, especially in that o f the medieval perspectivists, plays little role in this light was often assumed to be superior to the light that was diffused
what follows. Although the connections between light and theology are throughout the physical cosmos/’ Indeed, the empyrean was deemed to be
more fruitful, they are not our major source o f information for the relations invisible to us because o f opacity on its concave side, as if it were covered
between light and celestial bodies and hence are mentioned only briefly here by a curtain (cortina).~
at the beginning. The most fundamental cosmological link between light Theological works - hexaemeral treatises, commentaries on the Sentences,
and theology derives from the first chapter o f Genesis, which declares that
1. Bonaventure {Sentences, bk. 2, dist. 13], Opera, 1885, 2:311-329.
God created light on the first day and that he created the celestial luminaries
2. Thomas ot'Strasbourg reacted similarly. In his single but lengthy question on light, titled
to light heaven and earth on the fourth day. We would, therefore, expect “ Whether light is a real form” (An lumen sit torma realis), Thomas considers four sub­
to find discussions o f light in commentaries on the creation account o f themes, o f which only the first - whether celestial light is the same as terrestrial light - is
relevant, and this is treated briefly in less than half a column. The rest o f the question is
Genesis. During the Middle Ages, commentaries on the six days o f creation
given over to whether light is a real or intentional form; whether it is produced from the
were rarely made in special hexaemeral treatises, but they were usually potency o f a medium; and whether two intentional forms can be simultaneously in the
made in commentaries on the Sentences (Sentetitiae) o f Peter Lombard, same part o f a medium. See Thomas o f Strasbourg [SeHfei/a’s, bk. 2, dist. 13, qu. r], 1564,
15 Sr, col. 2-i_s6v, col. 1.
which, written around 1150, was the most famous theological textbook o f
3. See Thomas Aquinas [Sentences, bk. 2, dist. 13], 1929—1947, 2:328-340. The same applies
the Middle Ages and a required text tor comment by all theological students. to Aquinas's Sumina theologiae, which contains a series ot questions on the six days ot
There, in book 2, distinction 13, Peter discussed the first o f the six days of creation (pt. 1, qu. 67. arts. 1—4, 1967, 10:52-91).
4. As when Richard o f Middleton, in his commentary on the Sentences, bk. 2, dist. 13, art.
creation and asked whether light was corporeal or spiritual; w hy God found
1, qu. 4, 1591.2:1 59, col. 1, considered whether all light is o f the same species and concluded
it necessary to create the Sun if he had already created light on the first day; that celestial light differs from terrestrial light, since the latter “ is like a certain participation
and the meaning o f day and the distinction between day and night (Peter in celestial light, or an imitation o f it. ” Thus light trom terrestrial tires differs from the
light we receive from the celestial region.
Lombard, Sentences. bk. 2. dist. 13, ch. 2, 1971, 389-391). In book 2, 5. See Bonaventure, Sentences, bk. 2, dist. 14, pts. 1—2, Opera, 1885, 2:335-365. The same
distinction 14, Peter briefly mentions the creation o f Sun, Moon, and stars may be said for Thomas Aquinas (see Sentences, bk. 2, dist. 14, 1929—1947, 2:346-360 and
on the fourth day and asks about their utility (ibid., chs. 9-10, 398—399). Sumtna theologiae. pt. 1, qu. 70, arts. 1-3. 1967, 10:106-125).
6. Thomas Compton-Carleton [De coelo, disp. 2, sec. 3], 1649, 403, col. 2, made this dis­
For nearly hve centuries, scholastic commentators on the Sentetices rou­ tinction. He regarded empyrean light as “ a higher light than the light o f the other heavens
tinely included questions on the nature o f light. Most o f the questions, and o f the sublunar [region].”
however, discussed aspects oi light that were only tangentially, if at all, 7. Mastnus and Bellutus [De coelo, disp. 2, qu. 5], 1727, 3:506, col. 1, par. 157. Hurtado de
Mendoza [De coelo, disp. 2, sec. 5], 1615, 375, col. 1, par. 63, says the same thing. Arriaga
relevant to light as it affected celestial bodies. Saint Bonaventure, for ex­
[De caelo, disp. 1, sec. 6], 1632, 507, col. 2, says that although the empyrean heaven is
ample, inquired whether the light ot the first day was corporeal or spiritual; most lucid, it is commonly thought that it has an opaque cover on our side o f it. Although,
how light made dav and night; whether light is a body or the form ot a he adds, this iaim cannot be demonstrated, neither can it be rejected.

390
392 THE CELESTIAL REGION CELESTIAL LIGHT 393

and summas o f theology - are therefore not the best source o f information from which lumen arises” (Grant [Eastwood], 1974, 383). Although me­
about scholastic conceptions o f light in relationship to celestial bodies, al­ dieval authors may have “ ignored the distinction or employed it haphaz­
though one can learn about important aspects o f light itself: for example, ardly” (Bacon [Lindberg], 1983, 365, n. 10), it was repeated many times
that it was not a body but was either a substantial form o f a luminous body well into the seventeenth century.
- that is, the form that makes the Sun a luminous body" - or the accidental Another way to characterize the problem o f light sources is to view it as
quality o f a luminous body, so that light could be a quality derived from one in which natural philosophers sought to determine whether this or that
the substantial form o f the Sun,' or indeed both.10 One can also learn that celestial body, or group o f celestial bodies, was made visible to us by lux
opinion was divided about the transmission o f light, some assuming an or lumen.13
instantaneous transmission" while others argued that it was disseminated
successively, in very short but still temporal intervals.12 The most helpful
source o f discussion on celestial light is questions on De caelo. But even
here, the physical aspects o f celestial light were accorded only modest at­ III. A re the stars and planets self-lu m in o u s, or do th ey
tention (see the “ Catalog o f Q uestions,” Appendix I, qus. 226-234). The rece iv e their lig h t fro m the Sun?
most widely discussed question (qu. 231), and easily the most important, The role o f the Sun lay at the heart o f the problem. Was it the ultimate,
however, was whether all stars and planets receive their light from the Sun unique source o f celestial light for all planets and stars? If so, it might follow
or are self-luminous? Along with a few other aspects o f celestial light, the that all visible celestial bodies lacked light, or lux, o f their own and derived
source o f light for celestial bodies is the focus o f our attention here. their illumination from the lumen o f the Sun. According to Albert o f Saxony,
and many other scholastic authors, authorities were divided on this issue.
Aristotle and Averroes, who both assumed the Sun as the sole source o f
II. L u x and lu m en ligh t,14 were opposed to Macrobius and Avicenna, who “ concede that the
M oon has light from the Sun” but assume “ that all the other planets [and
In connection with light sources, scholastics followed Avicenna and distin­ stars] [astra] are self-luminous.” IS Lunar eclipses and the waxing and waning
guished two aspects o f light, to which they applied different terms. The
term lux was associated with light as the luminous quality o f a self-luminous 13. Mastnus and Bellutus, De coelo, disp. 2, qu. $, 1727, 3:50$, col. 1, par. 151, who give
body, such as, for example, the body o f the Sun. The light from a luminous much the same definitions, pose their question in the form: “ What are lux and lumen,
body that emanated into a surrounding medium, such as the Sun’s rays, and how do they occur in the heavens?” (Quid sint lux et lumen et quibus conveniant
in celis?). On p. 506, cols. 1-2, pars. 157-162, they discuss the theme we shall pursue in
was characterized by the term lumen (see Bacon [Lindberg], 1983, 365, n. the next paragraph.
10). The latter was thus capable o f lighting bodies that were not self- The term splendor was also used. In the thirteenth century, Thomas Aquinas used it
luminous. As Bartholomew the Englishman expressed it: “ Lumen is a certain (along with lux, lumen, and radius) to indicate reflected rays that reach a smooth and
polished surface, such as water and silver, and are then projected again (see Sentences, bk.
emission, or irradiation, by the substance o f lux. Lux is the original substance 2, dist. 13, qu. 1, art. 3, 1929-1947, 2:332). An anonymous fourteenth-century author
also employed splendor and defined the three terms as follows: lux is found in a lucid
8. Bonaventure has been cited as upholding this position (see Steneck, 1976. 47 and Wallace body, such as the Sun; lumen occurs in a transparent medium, such as air; and splendor
in Thomas Aquinas, Summa theologiae, 1967, 10:61, n. e). As we shall see, however, occurs in a body that has light by reflection. “ Thus indeed we say that lux is in the Sun.
Bonaventure seems to uphold both positions (see n. 10 o f this chapter). splendor in the other planets, and lumen is in the air.” (Dicimus quod lux est in sole,
9. This was the most popular view and was held, for example, by Thomas Aquinas (see splendor in aliis planetis et lumen in aere.) See “ Compendium o f Six Books," Bibliotheque
Summa theologiae, pt. i, qu. 67, art. 3, 1967, 10:61); Richard o f Middleton (Sentences, bk. Nationale, fonds Latin, MS. 6752. 214V. Although the term splendor was used to char­
2, dist. 13, art. 1, qu. 3), 1591. 158; and Henry o f Hesse (Steneck, 1976, 47). acterize light in the empyrean region, it was rarely used to describe the light ot planets.
10. Bonaventure held that both opinions were founded on some truth. It is a substantial form For further discussion o f these terms, see Conimbricenses [De coelo, bk. 2, ch. 7, qu. 9,
because insotar as other bodies participate more or less in lux, so do they have greater art. 2], 1598, 300-301.
or lesser being; but insofar as light is perceptible and is capable o f augmentation and 14. Plato, Timaeus, 39B, 1957, 115, also appears to have assumed that the Sun was the sole
diminution it is an accidental property. See Sentences, bk. 2, dist. 13, Opera, 1885, 2:321, source o f light for all planets and stars.
col. 1. 15. “ In hoc enim Aristoteles cum Averroi contrariantur Avicenne Macrobio and pluribus
11. This was the opinion o f Aristotle (De anima 2.7.418b.21-26), who was followed by aliis. Aristoteles enim in libro De proprietatibus elementorum vult dicere quod omnes stelle
Richard o f Middleton, Sentences, bk. 2, dist. 13, art. 2, qu. 2, 1591, 2:162, col. 1. Among habent lumen a sole, sicut luna et quod non sunt de se lucide quia apparet sibi quod
those who adopted this opinion in nontheological works were Robert Grosseteste (see consimili modo debeat esse de luna et aliis astns. Sed Avicenna, cum suis sociis. licet
Lindberg, 1976, 97) and Witelo (Lindberg’s translation in Grant, 1974, 395). concedat lunam habere lumen a sole, ponit tamen istam conclusionem: quod omnia alia
12. This opinion was incorporated into a theological work by Henry o f Hesse (see Steneck, astra habent lumen a se.” Albert o f Saxony [De celo, bk. 2, qu. 20], 1518. iif r , col. 2.
1976, 49). O f those who adopted this opinion, Alhazen was the chief authority and Roger See also Oresme [De celo, bk. 2, qu. 11], 1965, 640, who adds to Macrobius and Avicenna
Bacon one o f his leading followers (see Grant [Lindberg], 1974, 396-397, where Bacon the names o f Heraclitus and Tulius (Cicero).
also describes Alhazen’s opinion). Macrobius, in his Commentary on the Dream of Scipio, bk. 1, ch. 20 (1952, 168) adopts
THE CELESTIAL REGION CELESTIAL LIGHT 395
394

o f the Moon every month were sufficient to convince all that the Moon 1. The Sun as sole source o f celestial light
had its light from the Sun. During the Middle Ages most natural philos­
ophers seem to have sided with Aristotle and Averroes and argued for the In his discussion o f the Moon, Averroes provided one rather popular model
Sun as the sole source o f light for the stars and planets. But the answer was o f the Sun’s mode o f activity. The M oon derives its light from the Sun and
by no means obvious. Nicole Oresme and Albert o f Saxony insisted that is therefore not self-luminous. But “ it has been demonstrated,” Averroes
neither side o f the argument was d e m o n stra b le,alth o u g h both, for dif­ explains, “ that if the moon acquires the power ot lighting up from the sun,
ferent reasons, thought one side more acceptable than the other. Because it is not from reflection. That has been proven by Avenatha [that is, by
o f his “ love o f Aristotle, the Prince o f Philosophers,” Albert rejects six Abraham ben Meir ibn Ezra] in an interesting treatise.” 20 “ If it illuminates,”
arguments attributed to Avicenna and defends Aristotle’s opinion that all he continues, “ it is by becoming a luminous body itself. The sun renders
planets and fixed stars receive their light from the Su n.'7 B y contrast, O r­ it luminescent first, then the light emanates from it in the same way that
esme declares (near the end ot the question) that the self-luminosity option it emanates from the other stars; that is, an infinite multitude o f rays is
was more probable. '* Each side had a sweeping analogical argument based issued from each point o f the m oon.” 2' Duhem likened this process to
on two assumed truths: that the Sun shines by itself and that the Moon florescence and explained how Averroes used it to account for the dark
receives its light from the Sun. Those who thought the planets and stars spots on the Moon ([Ariew], 1985, 482): “ When the light ot the sun pre­
were self-luminous pointed to the self-luminosity o f the Sun. If the latter disposes and excites them, the various parts o f the moon become lumines­
was self-luminous, then, on the assumption that all celestial bodies are in cent; but they do not all become luminescent in the same w a y .” The parts
the same species, all other celestial bodies should be self-lum inous.19 B y the that become least luminescent show up as dark spots.
same token, those who viewed the Sun as sole source o f celestial light Albertus Magnus played a significant role in developing arguments in
reasoned that if the Moon received all its light from the Sun, then so also defense o f the Sun as the unique source ot celestial light. N ot only did his
should all other celestial bodies receive their light from the Sun. As we shall arguments exert an influence in the Middle Ages - Albert o f Saxony seems
now see, many representative arguments for these tw o opposing positions to have adopted his key ideas - but they continued to do so into the sev­
were ot this nature or were based on hierarchical or metaphysical consid­ enteenth century. Albertus’s treatment ot the subject thus provides a con­
erations. venient point o f departure for our discussion ot the Sun as the sole source
o f celestial light.22
That the Sun was the unique source o f celestial light was evident to
this interpretation from Cicero, who, according to Macrobius. allowed that the planets
receive light trom the Sun, but insisted that they “ also have their own light - that is, Albertus by a metaphysical argument that was frequently repeated well into
with the exception of the Moon, which, as we have repeatedly noted, is devoid o f light.” the seventeenth century. Something that exists in many things at once must
Macrobius was associated with this position well into the late sixteenth and the seventeenth exist primarily and fundamentally in one ot them, which tunctions as the
century, when he (and Avicenna) w-ere mentioned by the Conimbncenses, De coelo, bk.
2. ch. 7, qu. 4, 1598, 303, and by Riccioli, Almagestum novum, pars prior, bk. 6, ch. 2, cause o f all the rest. If light is an example o f this principle, as Albertus
1651, 393, col. i. who gives the reterences as book 1, chapter iy, but quotes the correct believed, then one thing must be the source and cause o f light in everything
passage. else. The Sun was the obvious candidate.23
16. Oresme, ibid, and Albert ot Saxony, ibid. Albert likens each o f these arguments to a
prohlema neutrum, that is. an argument in which two alternatives are either equally probable
or equally incapable ot demonstration (“ Breviter ista dubitatio utrum omnia astra prefer 20. Oresme, who accepted this argument (Le Livre Jit del, bk. 2. ch. 16, 1968, 457). explains
iunam et solem habeant lumen suum a sole est quasi unum probleuma neutrum, sic quod that reflection is not the mode o f diffusion ot the Moon's light. For if the Moon produced
rationes que hunt pro una parte possunt solui faciliter sicut rationes adducte pro alia” ); reflection like a mirror, “ the sun would appear in only a small portion of that part ot the
on probiemata neutra see Maier, 1949, 199. moon which seems lighted to us, and at times it would appear in no part at all. . . . It
17. “ Et ideo ob amorem Anstotelis. principis philosophorum. solvam rationes sex iam factas would be exactly as though we were looking at the sun in a mirror or in the water; we
pro opimone Avicenne contra Aristotelem tenendo cum Anstotele quod onines stelle do not see it from every position from which we can see the mirror, nor from every
preter lunam et solem, sive sint planete sive stelle fixe, habent lumina sua a sole." Albert angle, but only from a certain position and at a certain distance, and trom another distance
ot Saxony De celo. bk. 2. qu. 20, [$18, 115r. col. 2. Some ot these rejected opinions are we see it in another place.” All o f this follows, says Oresme, from the law o f reflection.
described below :n Section III.3b ot this chapter; see also Section 111.3a, for Albert’s 21. The translation is from Duhem [Ariew], 1985, 481. For the Latin text, see Averroes [De
important qualification at the end ot the question, where he allows that planets may have idelo. bk. 2. comment. 49], 1362-1574, 5: r3 1r. col. 2-131V, col. 1.
a weak light o f their own. 22. Albertus Magnus [De coelo. bk. 2, tract. 3. ch. 6 ;“ A digression explaining how all the
18. Oresme, De celo, bk. 2, qu. 11. 1965. 652. stars are illuminated by the Sun” )|, Opera. 19-1. 5. pt. 1:153, col. 1—155. col. 1. See also
19. Ibid., 644 (seventh argument). We saw earlier that most scholastics denied that all celestial pages 29 and 107, where Albertus discusses the dissemination o f celestial light.
bodies belonged to the same ultimate species. Those who used this argument would have 23. “ Omne quod est in muitis secundum un;im rationem. primo est in uno aliquo quod est
to adopt the single-species argument and then explain how all the diverse celestial bodies causa omnium illorum. sicut omnium caiidorum causa est ignis. Lumen ergo, quod est
could belong to a single species. multiplicatum in caelo et muitis modis est in lumimbus caeli, oportet, quod primo sit in
396 THE CELESTIAL REGION CELESTIAL LIGHT 397

Those who assumed this principle had to explain how the planets could The solar light that penetrates the planets is not reflected to us but is
appear visibly different and yet receive their light only from a single source, rather embodied or incorporated into them (incorporation est lumen stellis).
the Sun. Albertus coped with this problem by assuming that planets differ The light w e see in a planet is “ embodied light,” which lights up the planet
in their ability to receive the Sun’s light and that these differences were from its center to its circumference. Without sunlight, however, embodied
based upon a principle o f nobility: the greater the nobility o f a planet, the light cannot be retained continuously. It is not like the light ot a candle,
greater its capacity to receive solar light.24 In each o f the noblest and purest which burns continuously as long as it has matter to kindle it. When the
celestial bodies, light enters the surface on the side facing the Sun and moves matter is used up, the candle is corrupted and extinguished. If the planets
immediately to the opposite surface, thus filling the entire body with light, depended on some analogous matter to maintain their light, they would be
just like the light that burns in a candle.25 corrupted upon the exhaustion o f that matter. But planets and stars are
But the planets affect the light they receive in different ways. Thus Ju­ incorruptible and in this aspect are not analogous to candles. Rather, they
piter’s light has never diminished in purity from its first reception. By are like spherical vessels, which light up whenever a ray o f light touches
contrast, Mars, as a less noble body, alters the incoming light to a reddish them.
color, as does the star Aldebaran, whereas Venus converts its light to a pale The M oon is a partial exception. Albert o f Saxony, whose opinions on
hue and Saturn tends to obscure its light.26 Although these planets affect this are strikingly similar to those o f Albertus Magnus, who may have been
the light they receive in different ways, the Sun’s light penetrates them all his source, held that, like the other planets, the Moon not only received
and fills their interiors. O nly the M oon is an exception. Because it is o f a light directly from the Sun but embodied or incorporated that light into its
terrestrial nature, as Aristotle declared,27 the light received by the Moon transparent body. Indeed, Albert believed that the Moon was transparent
does not fill its entire interior but only penetrates partway. For this reason, at the surface and perhaps throughout its magnitude, although not as clear
the Moon, the least noble o f the planets, appears only partially illuminated as Venus and Mercury. Because o f its large size, however, sunlight was
and exhibits phases.28 unable to penetrate and fill the whole o f it. Consequently the side o f the
M oon turned away from the Sun is illuminated, albeit not as intensely as
uno, quod esc causa multitudinis hums. Ab illo autem quod est causa luminis, in omnibus
causatur lumen. Ergo oportet esse unum, a quo recipiatur lumen in omnibus quae lucent. the side turned toward the Sun.21' When we gaze upon the Moon, however,
Hoc autem nihil ita convementer ponitur sicut sol. Ergo a sole lumen est in omnibus the light “ we see is not only the light o f the Sun reflected on the body o f
stellis. Ibid., 153, col. 2, lines 83—93. In the fifteenth century, Johannes Versor and the M oon but [is also] the light o f the Sun embodied and incorporated
Thomas Bncot repeat and accept essentially the same argument (see Versor [De celo, bk.
2, qu. 11], 1493, 23V, col. 2, and Bncot [De celo, bk. 2], i486, 22r, col. 1 [“ Dubitatur [imbibitum et incorporation] into the M oon.” And like Albertus Magnus,
quinto utrum astrum aliud a sole lumen suum recipit a sole” ])- as did Johannes Velcurio Albert o f Saxony explains the variation o f certain properties and powers in
in the sixteenth century (Velcurio, [Physics], 1554, 76, col. 2). In the seventeenth century,
Aversa ([De caelo. qu. 35, sec. 3), 1627, 171, col. 2) and Amicus ([De caelo, tract. 6, qu. pt. 1:107, cols- I-2) thickness o f celestial bodies as the major criterion tor the reception
5, dubit. 1, art. 3], 1626, 359, col. 2) explicitly reject the argument. Because something o f light. There he declared that “ the cause o f light in stars is their thickness, and in some
is first in a genus does not signify, Amicus insists, that it is the cause o f the other, lesser the light [lumen] is received in their depths, and in others the light is diffused on the
things in that same genus. As the noblest member o f the genus, it only sets a standard surface. And those in which the light is received in the depths are made luminaries, like
for nobility, just as with colors, white is noblest but is not the effective cause o f other stars that light like candles; and those in which the light is diffused on the surface are
colors. made bright like milk, which is called the Milky Way . . . because there the orb is thicker.
24. For the same assumptions, see Versor, ibid., 23V, col. 1, and Bricot, ibid. And so it is necessary that the heaven be thicker and less thick so that its instrument,
25. “ Sed tamen in omnibus his tanta est pervietas, quod Stella recipiens lumen ex parte, qua which is light, can be diversified and move in different ways to different forms ot generated
vertitur ad solem, statim impletur ipso secundum totam superficiem et per omnia ir.ceriora and corrupted things.” In the later discussion that I described, Albertus ignores “ thickness"
eius, sicut si sit lumen accensum in candela."’ Albertus Magnus, De caelo, bk. 2, tract. 3, and stresses degree o f transparency, which seems associated with a body's nobility. Indeed,
ch. 6, 5 pt. 1:154. cols. 1—2, lines 43—47. Without mention ot a candle, Versor adopts the he even regards as false the claim that “ in order to [produce] light a pure transparent
same position (ibid.). thing requires only thickness” (Et quod dicitur, quod diaphanum purum ad hoc, ut luceat,
26. Albertus Magnus, ibid., 154, col. 1. It is obvious that Albertus has assigned a lesser non indiget, nisi quod sit spissum, falsum est). Ibid., bk. 2, tract. 3, ch. 6, 154, col. 2.
nobility to Saturn than to Jupiter, and thus implies, perhaps unwittingly, a rejection o f Noteworthy is Albertus’s apparent location o f the Milky Way in the celestial region
the concept that nobility was a direct function o f distance from the earth. among the stars, rather than in the upper reaches o f the sublunar region, as Aristotle had
27. In De caelo, bk. 2, comment. 42, 1562-1574, 5:i27r, col. 1, Averroes mentions that in insisted. Oresme seems to have adopted the same interpretation (see note 44 in this
his Liber de ammalibus Aristotle declares that “ the nature o f the Moon is similar to that chapter).
o f the nature o f the earth because o f the obscurity [or darkness] which is in it [that is. in 29. “ Unde dico quod lumen solis incorporatur in luna ita quod luna est corpus perspicuum
the Moon)."’ For a similar statement by Averroes, see his De caelo, comment. 49, 13 iv, et transparens, saltern circa superficiem eius et forte per totum, licet propter magnitudinem
col. 1. See Chapter 17, Section IV.3.a for further discussion. corporis lunaris lumen solis non possit totum corpus lunare penetrare sic quod eque intense
28. If the Moon were completely penetrated, as are the stars and other planets, it would appareat lumen in parte lune versa a sole sicut in parte lune versa ad solem." Albert ot
always appear fully illuminated as long as it received some sunlight, however little. Earlier Saxony, De celo, bk. 2, qu. 20, 1518, 1 i sr, col. 1. Thus “ thickness" - in the guise o f size
in his commentary, Albertus emphasized (De caelo, bk. 2, tract. 1, ch. 2, Opera, 1971, 5, or magnitude, rather than density - seems to have played a role for Albert o f Saxony.
THE CELEST IAL REGION C E L E S T IA L L IG H T 399
39«

different planets and stars by reason “ o f the different natures o f those stars that the M oon is not equally dense. Indeed, some parts o f it are so rare that
in which light is incorporated and embodied [incorporatur et imbibitur].',}° they cannot retain any light and are therefore observed as dark spots.33 But
The conception o f the planets as transparent bodies able to retain the light even in allowing rarity, which was usually associated with transparency,
o f the Sun - the sole source o f light in the celestial region - and to disseminate our author assumes that such rarity cannot hold light. Despite his assump­
it in proportion to their transparency may have become the most widely tion o f differing densities within the Moon, he insists that the Moon cannot
adopted explanation o f celestial light during the Middle Ages. 3 01 It continued hold sunlight but can only reflect it from its surface. Presumably this was
to have its supporters well into the seventeenth century.32 Because o f the also true o f all other planets. If so, the anonymous author had departed
medieval emphasis on nobility and hierarchy, it seemed natural to inquire drastically from the more widely accepted interpretations o f Albertus M ag­
w hy the Sun, as the source o f celestial light, was not placed above all the nus and Albert o f Saxony.30
planets and stars, so that its light could descend and illuminate all celestial In a surprising penultimate sentence to his question on solar light and its
bodies. Why did it occupy the middle, or fourth, position among the planets reception by the planets, Albert o f Saxony declares that the Moon and every
and light up both superior and interior planets, in violation o f the principle other star or planet have a weak self-luminous light that does not originate
that inferior bodies cannot influence or affect superior bodies? “ Wise nature” in the Sun. He infers this striking conclusion from observation o f a quarter
(.sapiens or sagax natura), responded Albertus Magnus, decreed that the planet M oon in clear air. Instead o f seeing only the quarter Moon, we see the
providing life-giving light and heat should be in the middle, so that it would entire face o f the Moon, although the part turned toward the Sun is more
be neither too far away, and thus leave bodies too cold, nor too close and lucid than that which is turned away from the Sun. If there were no light
cause bodies to become too hot.33 within the M oon itself, we ought to see only the quarter o f the Moon,
One problem, however, seems to have escaped explicit discussion. Why, because there are no solar rays hitting the part that is turned away from the
during a solar eclipse, should the side o f the Moon turned toward us be Sun.37 In his explanation o f this phenomenon, Albert concedes that “ the
completely darkened if the Moon is a transparent body? Would it not be M oon and every other planet have a weak and remiss light within them­
plausible to assume that sunlight striking the far side o f the M oon would selves. But that light which is noteworthy and which illuminates us per­
produce illumination sufficiently strong to be noticeable on the earth side ceptibly is from the Sun.” 3* In this way, Albert o f Saxony conceded some
o f the Moon? Perhaps it was with this in mind that an anonymous author element o f truth to the Avicennan position even as he supported the more
o f a fourteenth-century compendium o f natural philosophy adopted a rad­ traditional opinion o f the Sun as sole source o f light in the celestial region. 39
ically different position, arguing that all planets are opaque and dense bodies. Long before Albert o f Saxony’s acknowledgment o f a second source o f
As a dense and opaque body, the Moon can cause an eclipse o f the Sun planetary light, Richard o f Middleton had distinguished two kinds o f ce­
when interposed between us and the Sun.34*O n the very next page, however, lestial light in all the planets except the Sun. One kind is natural to the
in seeking to explain the spots on the Moon, he declares, inconsistently, planet itself but is weak and tenuous. Here we recognize the natural planetary
light which Albert o f Saxony also identified. The second light is not at all
30. The Latin text of the first o f these two passages is: “ Sic ergo lumen Iune quod videmus natural to the planets, because it is received from the Sun.40 From the
non est solum lumen solis reflexum super corpus lune sed lumen solis lune imbibitum et
incorporatum.” Albert o f Saxony, ibid. The second reads: “ Lumen unius asm causal
calorem et lumen alterius frigiditatem in istis mferioribus. Hoc est propter diversas naturas 35. "Compendium o fS ix Books,” MS. BN 6752, 22717 For the Latin text, see this volume.
ipsorum astrorum quibus illud lumen incorporatur et imbibitur.” Ibid., 115V, col. ;. Chapter 17. note 140.
31. For Oresme's description and acceptance o f this interpretation, see Le Livre du end. bk. t6. In his discussion o f lunar spots, he mentions Albertus Magnus and was probably familiar
2, ch. 16, 1968. 437-459, and Chapter 17, Section IV.3a.1i, o f this volume. In his de­ with his De cae'to.
scription o f this opinion. Duhem [Anew], 1985, 482, says that “ the property they at­ 37. At the outset o f the question, Albert says the following: “ Probatur primo de luna quod
tributed to the moon is not unlike what we call fluorescence.’’ habeat lumen ex se. Nam aere existente sereno et luna existente.in prima quadra videtur
32. For example, see George de Rhodes [De coelo. bk. 2, disp. 2, qu. 1, sec. 3], 1671. 284, totale corpus lune, licet ilia pars que est versa ad solem videatur lucidior quam alia pars
col. 2 (“ Dico secundo astra omnia, turn fixa, turn errantia, nullam videntur habere lucem versa a sole.” De caelo. bk. 2, qu. 20, 1518, 114V, col. 1.
propriam, sed ea omnia lluminuri a luce solis” ). 38. “ Ad primum bene conceditur quod luna et quodlibet aliud astrum habet lumen debile et
33. Albertus Magnus, De caclo. bk. 2. tract. 3, ch. 6, Opera. 1971, 5, pt. 1:154, col. 2. For remissum ex se. Sed quod habet lumen notabile, quod nos notabiliter llluminat. hoc est
the probable source o f Albertus’s opinion and for others who held it, see Chapter 11, a sole.” Ibid., u s v , col. i.
Section I.4 and note 35. For more on the central role o f the Sun, see Chapter 11, Section 39. In contrast to Albert, Thomas Bricot, De celo. bk. 2, i486, 22r, col. 1, rejects the idea
I.3 and especially note 28. that all planets except the Moon need not receive their light from the Sun.
34. “ Compendium o f Six Books,” Bibliotheque Nationale, fonds latin, MS. 6752, 226v. 40. “ Dicendum quod in luna et in qualibet alia Stella a sole, duplex est lux: una sibi naturalis.
Because the Sun is greater than the earth, our author observes that the Moon cannot quae debiiis est et tenuis; alia non sibi naturalis quam a sole recipit. Prima non est eiusdem
totally eclipse the Sun. For a discussion o f MS. BN 6752 and a description o f its contents, speciei cum luce solis; secunda tamen est eandem speciem reducitur cum luce solis.”
see Thorndike. 1923—1958, 3:568—584. For the Latin text, see Chapter 17, note 140. Richard o f Middleton, Sentences, bk. 2, disc. 13, art. 1, qu. 4, 1591, 2:159, col. 2.
400 THE CELESTIAL REGION CELESTIAL LIGHT 401

thirteenth century, at least two kinds o f planetary light were distinguished. “ so much more [should] the stars o f the heaven, which are more noble and
Although little was said about the light o f the fixed stars, it eventually ought to shine by themselves perpetually.” 4'’
emerged as a third distinct, though by no means weak, source o f light. Another argument trades on the nature o f sunlight, which was assumed
to produce heat and dryness. N ow , if the Sun’s light is the sole source o f
2. That the celestial bodies have some or all o f their light planetary light, reflected light from the planets should also produce heat
from themselves and dryness. But astronomers and experience reveal that “ some light from
the stars produces cold, and the light from the moon produces cold and
To arrive at a representative selection o f arguments that the planets and moisture.” 47 The inference is obvious: the planets and stars can produce
stars, with the exception o f the M oon, are partly or w holly self-luminous, such effects only by their own light, the effects o f which differ from sunlight.
we can do no better than draw upon Nicole Oresme and Albert o f Saxony, N ot to be ignored is the fundamental medieval argument that if the Sun
two o f the leading scholastic authors o f the fourteenth century.4' Both were shines by itself, the other planets ought to do likewise, because all are
convinced, as we saw earlier, that except with reference to the M oon, no members o f the same species.4S
demonstration could determine whether the Sun was the sole source o f light An argument by Oresme about the Moon resembles the one cited earlier
among the planets and stars.44 M ost noteworthy, perhaps, is the emphasis from Albert o f Saxony, except that where Albert speaks o f a phase o f the
(especially by Oresme) on arguments that appeal to the strong medieval M oon, Oresme appeals to a lunar eclipse, which, despite the exclusion o f
sense o f cosmic hierarchy and assume that things farther away from the solar light, manifests colors that are really lights, lights that could only
earth are “ better” and “ nobler” than those nearer to it. come from the M oon itself.49
The fixed stars ought to be self-illuminating because they are part o f the Supporters o f the Avicennan and Macrobian position argued that if all
primum mobile, or eighth sphere, which, according to Aristotle, is the most the planets had their light from the Sun and possessed none o f their own,
noble heaven.43 Since light is the most noble quality and the fixed stars the planets ought to exhibit variations in light - that is, they ought to undergo
most noble bodies in the most noble heaven, the fixed stars ought to shine phases - just as the M oon does, and some o f them ought to cause eclipses
by themselves.44 o f the Sun (Mercury and Venus) or be eclipsed by the earth (Mars, Jupiter,
Turning to the planets, Oresme explains that Saturn, Jupiter, and Mars and Saturn) when it is interposed between them and the Sun. But no such
ought to be self-luminous because they are superior to, or higher than, the phenomena are observed.50
Sun. This argument trades on an oft-repeated concept o f cosmic hierarchy
that superiors act on inferiors and not the contrary. Thus the Sun cannot
46. Oresme, ibid. Indeed, Oresme pushes this point even Farther in the next argument, when
act on the three superior planets and therefore furnishes no light to them.45 he explains that in the sublunar world things like fire and the scales ot certain fish are
Comets, Oresme observes, are self-luminous, shining by themselves at self-luminous; therefore, so much more ought the perfection o f self-luminosity to exist
night “ in the shadow o f the earth” in the elemental region o f the world in the far more excellent celestial bodies.
47. Ibid., 644. The “ experience” (or “ experiences” ) that would reveal information o f this
just below the Moon. Since comets can shine independently in this manner, sort is unmentioned. Albert o f Saxony presents a briefer version in De celo, bk. 2, qu.
20, 1518, 1 isr, col. 1.
41. In the thirteenth century, Bartholomew the Englishman. De proprietatilms rerum, bk. 8, 48. This argument would have had little force, because few believed that all the planets
ch. 30, 1601, 414, declared that the Moon “ has a substantial darkness, since it does not belonged to the same species.
have light from itself, as do the other planets” (Habet emm substantialem obscuritatem, 49. Oresme, De celo, bk. 2, qu. 11. 1963, 644.
cum non habeat sicut alii planetae lumen a seipsa). 50. Albert o f Saxony’s fourth argument (De celo, bk. 2, qu. 20, 1518, ii .sr, col. 2). Oresme
42. Although Albert believed that either side o f the issue was demonstrable, he sided with offers a similar argument in ibid., 640. but confines himself to Mercury and Venus. In
Aristotle and adopted much that Albertus Magnus espoused. the repudiation o f this argument, it was noted that no eclipses o f Mars, Jupiter, and Saturn
43. The primum mobile means the first movable sphere and was always the outermost moving could occur when the earth is interposed between them and the Sun, because the earth’s
sphere in the cosmos. Thus it could be any sphere from the eighth to the tenth. Since shadow does not extend to the planets above the Sun. As his authority for this, Albert
Aristotle identified the eighth sphere, or sphere o f fixed stars, with the primum mobile, invokes the Theorica planetarum o f Campanus o f Novara, who is alleged to have said that
Oresme merely repeats the identification. the earth’s shadow does not extend beyond Mercury (Albert, ibid., 115V, col. 1). But
44. Oresme, De celo, bk. 2, qu. 11, 1965, 640-642. In the third argument, Oresme identifies even if it did. Mercury remains so close to the Sun that it the earth’s shadow did reach
the nature o f the Milky Way with that o f the eighth sphere and argues that the Milky it, the earth's diameter would not fit between Mercury and the Sun and therefore could
Way must also shine by itself. Oresme seems to have located the Milky Way in the not cause an eclipse.
celestial region, as did Albertus Magnus (see this chapter, note 28), which marks a sharp Actually, the data Campanus provides indicate that the earth's shadow would extend
departure from Aristotle (see Chapter 14, Section VIII. 1. b.i and Chapter 19, Section III.2 beyond Mercury and reach as far as the sphere o f Venus. Thus the apex ot the earth's
and note 26, for Aristotle’s 'dews). shadow extends 866.536 4/11 miles from the earth’s surface to the apex ot its cone
45. Oresme, ibid., 642. Albert o f Saxony's second argument in defense o f the Avicennan (Campanus, Theorica planetarum, 1971. 149). In summary tables for Mercury and Venus,
position is basically the same (see his De caelo, bk. 2, qu. 20. 1518, 1 x5r, col. 2). For the concave surface o f the sphere o f Mercury is 209,198 miles and its convex surface
Roger Bacon as a possible exception, see Chapter 19, note 5. 579,320 miles (ibid., 358). Obviously, the earth’s shadow will extend beyond this and
402 THE CELESTIAL REGION CELESTIAL LIGHT 403

Finally, the lights o f the planets differ from one another: Mars is like the As we saw, however, Albert o f Saxony had already conceded that solar
light ot tire, and Saturn is like that o f white lead. But if they all have their light was not the only source o f planetary and stellar illumination: to a very
light from the Sun, they ought to be o f similar appearance.51 small extent, the Moon, and by inference the rest o f the planets, seem to
be weakly self-luminous. Unwittingly, Albert adumbrated an interpretation
that would become popular in the sixteenth and seventeenth centuries. The
j. Seventeenth-century scholastic interpretations and the Coim bra Jesuits bore witness to this when, at the end o f the sixteenth
new discoveries century, they declared: “ the more common assertion o f the astronomers is
that both the fixed stars and the planets receive their light from the Sun but
We must emphasize that the seemingly most popular explanation o f pla­
nevertheless possess some light by themselves.” 53 Am ong scholastic sup­
netary light - the one adopted by Albertus Magnus and Albert o f Saxony
porters o f this interpretation in the seventeenth century were Bartholomew
- did not explain the visible light o f the planets in terms o f the direct
Amicus, Thomas Compton-Carleton, and Melchior Cornaeus.54 All would
reflection o f solar light from their allegedly opaque bodies. Rather, it as­
have agreed, however, that the Sun provides the greater part o f planetary
sumed planets and stars that were transparent and could therefore receive
and stellar light. 55 Their interpretation assumed that most o f the light for
solar light throughout the extent o f their bodies. They were visible to us
celestial bodies came from the Sun but that all o f these bodies also had some
because they were completely filled with the Sun’s light. During the Middle
additional source o f light from within themselves.
Ages, this interpretation was opposed by one attributed to Avicenna, which,
Another rather widely held opinion was opposed to the one described
as we saw, assumed self-luminosity for every planet and star except the
by the Coim bra Jesuits. The rival interpretation either explicitly or by
Moon. Both o f these rival opinions were encapsulated by Nicholas C op­
implication denied any degree o f self-luminosity to the planets, including
ernicus, who speaks o f those who “ do not admit that these heavenly bodies
the Moon. All planetary light derived directly from the Sun and was sub­
have any opacity like the m oon’s. On the contrary, these shine either with
sequently disseminated by reflection. Within this group, the planets could
their own light or with the sunlight absorbed throughout their bodies.” 52
be viewed as entirely opaque bodies or as materially different bodies that
were unequally lucid and possessed different degrees o f purity. Johannes
into the orb ot Venus, the concave surface ot which is 579.320 miles and its convex
surface 3,892.866 mile' ibid., 359). Velcurio was in the latter group,5'’ while Raphael A versa and George de
In the seventeenth century, Melchior Cornaeus (De coelo, tract. 4, disp. 2, qu. 4, dub. Rhodes are identified with the former.'
9, 1657, 512) explained that the fixed stars, which receive their light from the Sun, were Let us now consider seventeenth-century opinions on the Moon, the other
not eclipsed by the earth s shadow' because the latter culminated in the apex of a cone
and never reached the firmament. Thomas Compton-Carleton, De coelo, disp. 2, sec. 3, planets, and the fixed stars, in that order.
1649, 403, col. 1, says much the same thing as Albert ot Saxonv, denying that Mercury
and Venus could eclipse the Sun and that the earth can eclipse the superior planets. He 53. “ Statuenda camen est hec assertio quam communior astrologorum consensus approbat:
explains that Venus is only a hundredth part of'the Sun (in volume?) and Mercury is even tam stcllae fixae quam planetae lumen a sole mutuantur, ita tamen ut aliquid ex se lucis
smaller; and since the Sun is 166 times greater than the earth (presumably in volume), possideant.” Conimbricenses, De coelo, bk. 2, ch. 7, qu. 4, art. 1, 1598, 303.
the earth’s shadow cannot reach the superior planets. Drawing upon “ Albatemus, Thebit, 54. In the first o f five conclusions about the source o f stellar and planetary light. Amicus
and other astronomers,” Galileo ([De caeio, qu. 2 (H)J, 1977, 79), explains that “ the visual declares (De coelo, tract. 6, qu. 5, dubit. 1. art. 3. 1626, 358, col. 1) that “ the stars [and
diameter ot the sun has a tenfold ratio to the visual diameter o f Venus,” so that geo­ planets] have some light from themselves.” Compton-Carleton (De coelo. disp. 2. sec. 3.
metrically “ the visual diameter ot the sun . . . has a hundredfold ratio to the visual circle 1649, 403. col. 1) assumes that “ Neither the Moon, nor the other planets, nor even the
ot Venus.” Van Helden, 198s, 71. savs this relationship is found in Ptolemv's Pluretory fixed stars, receive all their light from the Sun” (Nec luna aut alij planetae. nec stellae
Hypotheses etiam fixae omnem iucem suam accipiunt a sole). In his reply to the question “ Whether
51. Albert ot Saxony's sixth argument (De celo, bk. 2. qu. 20. 1518, 11 sr. col. 2). Earlier in stars receive light from the Sun" (An stellae accipiant lumen a sole), Cornaeus, De coelo.
this section, we saw that Albertus Magnus replied to this argument by insisting that the tract. 4, disp. 2. qu. 4, dub. 9, 1657, 512, responds: “ Yes, but nevertheless they also have
planets had different capabilities for receiving the Sun's light and that these differences some from themselves” (Ita, sed tamen aliquod etiam ex sese habent).
produced different colors and appearances. 55. Compton-Carleton, ibid., declares that “ the stars [and planets] receive the greatest part
52. Copernicus, Revolutions, bk. 1, ch. 10 [Rosen], 1978, 19. Copernicus adds (ibid.) that o f their light from the Sun. This is manifestly proved from eclipses o f the Moon and also
these planets “ do not eclipse the sun, because it rarelv happens that thev interfere with from the fact that planets nearer to the Sun are illuminated more and more intensely, as
our view o f the sun, since they generally deviate in latitude. Besides, they are tiny bodies if they were ablaze, as is obvious in Mars and Venus."
in comparison with the sun. Venus, although bigger than Mercury, can occult barely a 56. Velcurio declares (Physics, 1554, 76, col. 2-77, col. i): “ Ergo omne lumen svderum refertur
hundredth of the sun. So says Al-Battani o f Raqqa, who thinks that he sun's diameter ad solena. . . . Ergo sol est causa efficiens luminis in omnibus stellis. Et per consequens
is ten times larger [than Venus s], and therefore so minute a speck is not easilv descried omnes stellae recipiunt lumen suum a sole." He goes on to explain: "Q uod autem non
in the most brilliant light.” In this passage, Copernicus epitomizes widely held medieval omnia astra sunt aeque lucida, neque eiusdem vel paris luminositatis, nihil impedit haec
views, especiaiiv that Venus “ can occult barely a hundredth o f the sun." Copernicus nam culpa non soils est. sed materia in Stella quarum quo quaeque est punor et nobilior.
immediately rejects this interpretation. That Venus could only occult a hundredth o f the . . . Quo autem quaeque spissior est. . . . Et quo iusto rarior est.”
Sun was repeated by Compton-Carleton, De coelo, disp. 2. sec. 3, 1649, 403, col. 1. 57. Despite divergence on important points. Aversa and de Rhodes were influenced by Galileo.
404 THE CELESTIAL REGION CELESTIAL LIGHT 405

a. The Moon M oon is perfectly opaque and in no way transparent. Solar rays are incapable
o f penetrating below the lunar surface.62 B y contrast, Aversa’s contem­
The problem o f the M oon’s source o f light continued to loom large in the porary, Bartholomew Amicus, accepted the opinions o f Albertus Magnus
seventeenth century. The problem was essentially the same as it was for and Albert o f Saxony — citing both in the course o f his rather lengthy
Albert o f Saxony: how to explain light or colors in areas o f the M oon in discussion - and assumed that the Moon (as well as all the other planets)
which solar rays were absent and which should have been totally dark and had light o f its own which it receives from the Sun and “ incorporates”
invisible. Raphael Aversa enunciated the problem succinctly: H ow can we (incorporatio) into itself, in the manner described b.y his medieval predeces­
explain a certain whiteness in that part o f the Moon when there is no Sun sors.6' The illumination o f the Moon could not be explained solely in terms
shining on it, and how, during the course o f an eclipse, do we account for o f reflection, since only a small part o f the Moon would be seen as illu­
a certain ruddy color?5* Although Albert o f Saxony and others in the Middle minated, namely that observed part where the angle o f incidence equals the
Ages had discerned lights in the darkened areas o f the Moon, Aversa was angle o f reflection, as is obvious with mirrors. If the Moon diffused light
more than likely summarizing Galileo’s distinction in the Sidereus nuncius solely by means o f reflection, only a small part o f the earth would be
between tw o different kinds o f lunar light — one in the normal course o f illuminated. Amicus concludes that the Moon receives solar light into its
the changing dark and light areas o f the lunar surface and the other detected very depths, or, as he put it earlier, “ incorporates” the light o f the Sun into
during lunar eclipses.559 Aversa explains that “ there is doubt from whence
8 itself and then diffuses it to the earth.64
and how the Moon«has this light, whether from itself or from the same If Aversa found that the Moon was not diaphanous and did not receive
Sun; and if from the Sun, whether it is directly from the Sun or from solar light within itself, he was equally convinced that it lacked an innate
another body illuminated by the Sun.” 60 light o f its own. Such a light should be visible throughout the M oon’s
Am ong responses to such questions was the Albertus M agnus-Albert o f circuit around the earth, which is manifestly not so.65 Nor does it receive
Saxony medieval explanation in which the Moon is conceived as somewhat additional weak light from Venus, as some, including Tycho Brahe, be­
transparent rather than absolutely opaque. The M oon is thus assumed per­ lieved/16
vious to solar rays and capable o f receiving a weak light that penetrates
lumen a sole. Ita opinati sunt illi. qui supenon quaest.. sect. 2. dicebant lunam non esse
beneath its surface.6' Aversa rejects this interpretation, insisting that the perfecte opacam, sed esse aliquantum diaphanam et ita perviam radijs eiusdem solis.”
Aversa, De caelo, qu. 35, sec. 2, 1627, 168, col. 2. This light, which is ultimately from
58. Amicus describes a similar phenomenon when he proposes the Moon as evidence that the Sun, should be distinguished from the weak, self-luminous lunar light that Albert o f
the planets had some light o f their own. Although the Moon seems to receive all its light Saxony mentions.
trom the Sun, it nevertheless seems to have some light o f its own, because it becomes 62. “ Verum satis ibi ostensum est lunam esse potius ita perfecte opacam ut transitum non
reddish during an eclipse, which could only occur if it had some light within itself (Primo praebeat in sua profunditate radijs solis.” Ibid.
patet ex luna quae videtur totum a sole lumen recipere et tamen habet aliquid luminis ex 63. As he so frequently does, Amicus, De caelo, tract. 6, qu. 5, dubit. 1, art. 3, 1626, 360,
se, ut patet in ipsius eclypsi, cum tunc rubeat, qui rubor non posset esse nisi aliquid col. 1, first poses a question in the name o f an imaginary opponent or critic: “ You ask
luminis in se construaret. Ergo idem dicendum de alijs astris). Amicus, De caelo, tract. [literally “ sav” ] how does the Moon receive light from the Sun” ? (Dices quomodo luna
6, qu. 5, dubit. 1, art. 3, 1626, 358, col. 1. Thomas Compton-Carleton repeats the same recipiat lumen a sole?) “ I reply,” says Amicus (literally “ Let it be replied” ) that “ it receives
argument (De coelo, disp. 2. sec. 3, 1649, 403, col. 1). Amicus’s opinion is similar to Albert it by incorporation to the extent o f a one-third part o f itself” (Respondetur recipere illud
ot Saxony’s in behalf ot a small amount o f weak light in the Moon itself. Indeed, Amicus per incorporationem. usque ad tertiam eius partem).
draws on a number ot arguments which Albert o f Saxony presented in defense o f the 64. "Quia si non per incorporationem esset per reflexionem. Unde quando luna esset ad latus
self-luminous nature o f celestial bodies. nostn aspectus, non appararet illuminata nisi parva pars lunae, scilicet pars a qua angulus
59. Gaiilec, Starry Messenger (Sidereus nuncius) in Drake. 1957, 42-43 and Galileo, Sidereus incidentiae et angulus reflexionis ad nostrum aspectum essent aequales. Turn quia si esset
nuncius [Van Helden), 1989, 54-55. solum corpus reflectens non posset illuminare nisi tantam partem terrae quanta est in se,
60. The full text follows: “ At vero quia visu ipso cernimus partem quoque lunae avertam a ut patet ex reflexione speculorum. Et tamen apparet totam terram illuminari que est maior
sole quando luna non est plena et videmus illam albicantem; imo et in eclipsi conspicimus quam luna est. Ergo dicendum est recipi in profunditate et inde diffundi lumen.” Ibid.,
adhuc lunam tusco quodam Iumine rubescentem. Dubium est unde et quomodo luna hoc 360, cols. 1-2.
lumen habeat: an a se vel ab eodem sole; et si a sole, an ab illo immediate vel ab alio 65. Aversa, De caelo, qu. 35, sec. 2, 1627, 169, col. 1.
corpore illuminato a sole.” Aversa, De caelo, qu. 35, sec. 2, 1627, 168, col. 2. Mastrius 66. These opinions were all briefly mentioned by Galileo in the Sidereus nuncius. In four or
and Bellutus asked similar questions (De coelo, disp. 2, qu. 5, 1727, 3:506, col. 1, par. five lines Galileo describes them all (Galileo, Sidereus nuncius [Van Helden], 1989, 54).
157) about all the celestial bodies, including the Moon: “ In another part o f what is sought, Thus in speaking o f the white light that shines even in the darkened parts o f the Moon,
we ask in what celestial bodies does light [/m.y ] occur: whether all the heavens and celestial Galileo comments: “ Some have said that it is the intrinsic and natural brightness o f the
bodies [astra] have light, or only the celestial bodies. And o f these [i.e., the celestial bodies] Moon herself; others that it is imparted to it by Venus, or by all the stars; and yet others
whether all, or only some have light” (In altera parte quaesiti quaeritur quibus corporibus have said that it is imparted by the Sun who penetrates the Moon’s vast mass with his
coelestibus conveniat lux: an scilicet omnes coeli et astra habeant lucem an solum astra; rays. But such inventions are refuted with little difficulty and demonstrated to be false.”
et istorum an omnia, an vero quaedam tantum). Galileo rejected the Venus explanation as “ childish,” saying “ For who is so ignorant as
61. “ Primus dicendi modus est lunam non esse penitus opacam et imperviam lumini, sed not to know that near conjunction and within the sextile aspect it is entirely impossible
esse aliquantum diaphanam et ita etiam in profunditate recipere immediate debile saltern for the part o f the Moon turned away from the Sun to be seen from Venus?” See also
406 THE CELESTIAL REGION CELESTIAL LIGHT 407

As a scholastic aware o f current opinion, A versa also considers what he the earth would not reach the Moon because they are too weak to reach
calls Galileo’s “ very new mode o f explanation’’ (valde novus dicendi modus beyond the second region o f the air.'0
est Galilaei). Although the Moon receives almost all o f its light from the
Sun, Galileo argued, in the Sidercus nuticius, that reflections o f sunlight from
the earth to the Moon augmented the M oon’s light and affected otherwise b. The other planets
darkened areas o f the lunar surface, a phenomenon Galileo called “ secondary Due in large measure to Galileo’s observations with the telescope, the range
light.” Galileo’s reasoning, Aversa explains, is simply that if the smaller o f explanations about the source o f celestial light was considerably expanded
Moon can illuminate the much bigger earth, it follows that the earth, which in the seventeenth century. B y his discovery o f the phases o f Venus in
receives more solar light than the M oon, should reflect some o f it to the 1610,71 Galileo had shown that another planet besides the M oon revealed a
lunar surface. Aversa finds the consequences o f this belief unacceptable, continual variation in the amounts o f lightness and darkness on its surface
since it inverts the order o f the universe/’7 that is, instead o f the heavens and did so as a consequence o f its position in relationship to the Sun. It
sending down light and other influences to affect the terrestrial region, as was not only convincing evidence that Venus revolved around the Sun
everybody believed, Galileo's explanation reverses the process and has the rather than the earth but that the brightness o f Venus was produced by the
earth, and therefore the terrestrial region, affecting and influencing the no­ Sun’s light. Since Galileo had shown that darkness lay over that part o f the
bler celestial region. Aversa expressed further displeasure at Galileo’s in­ Venusian surface where the Sun failed to shine, it was implausible to suppose
clusion o f the earth among the planets, a move that implied that the planets that Venus had its own independent source o f light. Without solar light, it
are like our earth, an association that was repugnant to defenders o f tra­ would lie immersed in darkness. On this approach, the planets were best
ditional cosmology. “ The earth,” Aversa insists, “ is an unpolished, uneven, interpreted as opaque rather than transparent bodies. Moreover, it Venus’s
and rough body, [whereas] the Moon is just like every other planet: it is brightness was derived from solar light, it seemed a further reasonable
very pure and very polished, more like out metals and gem s.” OH inference that solar light was also the cause o f the varying degrees o f color
From whence, then, does the M oon’s secondary light derive? It comes and brightness in the other planets. A powerful reinforcement to the phases
from the surrounding parts o f the heavens, seemingly from sunlight re­ o f Venus was Galileo's slightly earlier observation and description o f the
flected to the Moon from other celestial bodies. Obviously this excludes Moons, or satellites, o f Jupiter." With these powerful claims, scholastic
the earth, which for Aversa is not a planet. In a curious sense, Aversa natural philosophers in the seventeenth century faced a significant challenge
accepted an argument resembling Galileo's, but instead o f the earth as the to traditional interpretations about the planetary sources ot light.
M oon’s source o f reflected sunlight, he vaguely invoked other neighboring O f the serious opinions offered as explanations for the source or sources
celestial bodies. Mastrius and Bellutus adopted the same explanation and o f planetary light, none denied the role o f the Sun. In general, three opinions
even mentioned Aversa and three other seventeenth-century scholastic au­
thors (Licetus, Scheiner, and Tanner) w ho also accepted it. Speaking ex­ quern afferunt Aversa, quaest. 35, sect. 2: Licetus lib. de Lap. Bon., cap. 50; Scheiner disp.
plicitly o f “ secondary light” (lumen secundarium), they trace its source to 37; Tannerus I disp. 6, quaest. 7 et alii, lumen illud esse quoddam lumen secundarium
“ the nearer parts o f the heaven illuminated by the Sun.” To counter Galileo, proveniens a partibus celi propinquionbus a sole illuminatis. . . . Turn quia radii solares
per reflexionetn a terra non transcendunt secundam aeris regionem, aliter haec esset
they argued that if solar light were reflected from the earth and reached the calida. ” The second, or middle, region of air is the location o f clouds and is always cold
second region o f the air on its way toward the Moon, it would heat the (see Ch. 20 [immediately preceding Sec. I] for d’A illvs description o f the sublunar region!.
intervening air.6" This does not occur. Moreover, solar rays reflected from 70. “ Videmus enim reflexionem radiorum soils a terra non pertingere secundam aeris regi­
onem, nam reflexio non diffunditur ad quameunque distantiam propter imbecilhtatem
specierum.” Mastrius and Bellutus, ibid.. 507, col. 1. par. 162. In the preceding sentence
the translation in Galileo. Starry Messenger [Drake], 1957, 42-43. For references to Tycho, (also see note 69), Mastrius and Bellutus say that if sunlight reflected from the earth were
see Galileo, Sidereus mmcius [Van Helden], 1989, 54, n. 52. As we shall see, Aversa seems sent toward the Moon it would heat the air o f the "second region.” They also say that
to have adopted as his explanation a version o f the interpretation that attributed the cause the reflected ravs are too weak to reach the Moon. Would those rays also be too weak
o f the Moon’s extra light to “ all the stars.’’ to heat the air?
67. “ Verum de se satis incredibile apparet immitti et diffundi lumen a terra ad caelum, ad 71. In his First Letter on Sunspots to Mark Welser, dated May 4, 1612, Galileo expresses
caelestia corpora, quod plane est invertere ordinem umversi.” Aversa, De caelo. qu. 35, astonishment that his opponent Apelles (Christopher Scheiner) had not yet heard o f his
sec. 2, 1627, 170. col. 1. method for determining whether “ Venus and Mercury revolve about the sun or between
68. “ Nec unquam obtinebit Galilaeus, ut terra credatur unum ex astris et ullum astrum the earth and the sun.” Galileo explains that this method was "discovered by me about
credatur ut terra. Terra est corpus rude, asperum, et ruidum [in place o f ruuidum]; luna, two vears ago and communicated to so many people that by now it has become notorious.
sicut et omne aliud astrum, est corpus summe tersum et perpolitum plusquam apud nos This is the fact that Venus changes shape precisely as does the moon.” Galileo, Letters on
metalla et gemmae.” Ibid. Sunspots (1613) in Drake, 1957, 93. The discovery was made in 1610 but after publication
69. After citing Galileo’s explanation o f secondary light, Mastrius and Bellutus declare {De o f The Starry Messenger in March o f that year.
coelo, disp. 2, qu. s, 1727, 3:506, c°l- 2> par. 159): “ Sed probabilior dicendi modus est 72. In his Sidereus nuticius ot March 1610.
408 THE CELESTIAL REGION CELESTIAL LIGHT 409

emerged. The first assumed that the Sun is the sole source o f planetary extent self-luminous.76 Upon inspection o f the arguments, which are oc­
light; this opinion has three subdivisions. In one, the planets are assumed casionally ambiguous, the light peculiar to a planet is either solar light that
opaque and their light is but a reflection o f solar light. In another, the planets is received within a diaphanous planet or a kind o f light that is left uniden­
are assumed partly or wholly transparent and, instead o f reflecting solar tified. In those instances where the light o f the planet is transformed solar
light, the latter penetrates the diaphanous planetary matter and fills it with light, Albertus Magnus and/or Albert o f Saxony is probably the ultimate
light, thus rendering the planet visible because it is always filled with light. source. For example, Johannes Poncius, the seventeenth-century Scotistic
In the third opinion, for which Campanus o f Novara apparently qualifies, commentator, declares that “ planets receive their light principally from the
the Sun is assumed the source o f planetary light, but no judgm ent is made Sun,” 77 offering as reasons the varying brightness o f planets in proportion
about the opacity or transparency o f the planets.73 to their distance from the Sun and the middle position o f the Sun with
The second opinion attributes the primary source o f planetary light (lu­ respect to the other planets, which made it easier to communicate light to
men) to the Sun but also assumes that each planet has within itself a source them. Indeed, if the M oon is self-luminous, it ought to appear lucid during
o f light (lux). The third opinion is the now familiar one, derived from a lunar eclipse.
Macrobius and Avicenna, that only the M oon receives light from the Sun Poncius then qualifies his statement that the planets receive their light
but all other planets are self-luminous. Although this opinion was frequently “ principally” (principaliter) from the Sun, “ because it is not improbable that
repeated in the Middle Ages, few, if any, accepted it. In the seventeenth they have some light from themselves.” As evidence o f this, he mentions
century, Aversa indudes it in his discussion, even mentioning Macrobius only the M oon and gives a common instance: the Moon reveals some light
and Avicenna by name, and cites a few o f its alleged supporters (Molina, during its eclipse. B y the time Poncius wrote, the lunar light seen during
Lucillus Philalthaeus, and Scaliger). an eclipse and even light seen in the darkened portions o f the lunar surface
Those who assumed that the Sun was the sole source o f planetary light during its normal phases were usually explained as the result o f solar light
and that it shone on opaque planets could appeal for support to Galileo’s reflected to the M oon from one or more other celestial bodies, and even
telescopic observations, as did Raphael Aversa. Citing Galileo’s Letter on the earth (as Galileo argued). That was as far as Poncius would go with
Sunspots, Aversa invokes Galileo’s telescopic observations o f the phases o f claims for self-luminosity. Indeed, he rejected a general claim for it. He did
Venus and concludes that “ according to this observation it is clearly dem­ this by posing a comm on objection against his position that the Sun is the
onstrated that Venus does not have native light but is illuminated by solar principal source o f planetary light: what o f the frequently mentioned ar­
light.” 74 A paragraph later, Aversa introduces the satellites, or Moons, o f gument that even when the earth is interposed between the Sun and other
Jupiter and mentions that one or another o f them is always eclipsed by the planets, the latter are not eclipsed but continue to shine brightly? Is this not
shadow o f Jupiter when the latter is interposed between its satellite and the evidence o f self-luminosity? O n this piece o f evidence, Poncius gives the
Sun. Aversa agrees that Jupiter and its satellites receive their light from the usual negative response: the earth's shadow reaches only to Venus and
Sun. Indeed, he suggests that this may also explain the behavior o f the two therefore cannot cause an eclipse o f Saturn, Jupiter, and Mars, which con­
“ satellites” allegedly observed around Saturn.75 O n the basis o f such pow ­ tinue to shine from the reception o f solar light. And although the earth’s
erful evidence, Aversa seems to have accepted the idea that all the planets shadow can reach Venus and Mercury, those two planets do not recede
receive their light from the Sun and have none o f their own. sufficiently from the Sun to enable the earth to be interposed between
In this Aversa may have been unusual. Alm ost all other seventeenth- them.7*
century scholastic natural philosophers accepted the Sun as the primary Mastrius and Bellutus also argued for some sense o f self-luminosity,
source o f light for the planets but also assumed that the latter were to some commencing their discussion with an assertion that the other planets (astra)
are probably “ illuminated not only by the Sun but also have a proper greater
73. I interpret Campanus, Theorica planetarum, 1971, 148, lines 79-88 (translation on p. 149)
light, [just] as [does] the whiteness o f the M oon.” 79 Like most scholastics,
in this manner.
74. After citing “ Galilaeus in Epist. 3, de Macuiis Solis,” Aversa (De caelo. qu. 35, sec. 3, 76. We saw earlier that Amicus, Compton-Carleton, and Cornaeus held such an opinion.
1627, 172, col. 2) goes on to describe the opinion o f Galileo and likeminded supporters: To them we may add Poncius and also Mastrius and Bellutus.
“ Docent planetas illummari a sole quia iam per telescopium certis observationibus de- 77. “ Planetae suam lucem principaliter a sole participant.” Poncius [De coelo, disp. 22, qu.
prehensum est Venerem mutari ac variari ad instar lunae et non in suo orbe plenam sed 8], 1672, 62s, col. 1.
vere comiculatam apparere ita ut qua parte versa est ad solem notabili lumine fulgeat et 78. Ibid. The same opinion was held in the Middle Ages (Campanus o f Novara, Albert o f
in oppositum cornua vertat. Iuxta hanc ergo obervationem perspicue demonstrator Ve­ Saxony) and by others in the seventeenth century (Cornaeus and Compton-Carleton).
nerem non suo nativo lumine, sed solari illustratione lucere. ” See note 30. this chapter.
75. The rings o f Saturn were initially interpreted as satellites. 79. “ Tandem de aliis astris dicendum probabiliter videtur, non solum a sole illuminari, sed
4i o THE CELESTIAL REGION CELESTIAL LIGHT 41 I
they insisted that the Sun was the primary source o f planetary light, ap­ In the course o f six arguments intended to illustrate that planets (astra)
pealing, as did A versa, to the phases o f Venus, Mercury, and even Mars have their own light, Amicus provides no real clue as to the nature o f that
to buttress their claim.80 But what about self-luminosity? They give two light, although it is clear that he regarded the Sun as the primary source o f
reasons for accepting it. First, the planets (astra) must be diaphanous, for planetary light.H3 The kinds o f arguments he presents are unhelpful in de­
otherwise they would not even be seen. For their second reason they men­ termining the nature o f the proper light allegedly possessed by a planet.
tion the ditferent colors o f the planets, a phenomenon that indicates “ a Thus in his fifth argument. Amicus uses an analogy with the Sun. Since
difference o f light, and yet if they were illuminated by the Sun, the light the Sun is lucid, the other planets should also be lucid, because they are o f
in all o f them ought to be o f the same kind.” 8' Taken together, these tw o the same generic, though not specific, nature. By their single generic nature
reasons strongly suggest that Mastnus and Bellutus had in mind two man­ they should all have light o f their own, but by their different specific natures
ifestations o f light: sunlight and another kind o f light that is peculiar to each the degree and intensity o f light varies in each.84 Amicus does not say
planet. Both are explicable in terms o f the medieval interpretations asso­ whether this generic light, which is specifically adapted to each planet, is
ciated with Albertus Magnus and Albert o f Saxony. The diaphanous nature the Sun’s light. That planetary light differs from sunlight is conveyed by
ot each planet enables it to receive the Sun’s light throughout its body. Amicus in his third argument. Because planets (astra) exercise different
Indeed, sunlight fills it. It is able to shine because it then diffuses its sunlight. effects on inferior, sublunar things, they cannot act only by means o f sun­
Each planet absorbs light in its own way and thus alters it. These alterations light. For if the planets received only sunlight, it would be identical in each
ot the light are manifested by the different colors o f the planets. o f them, and they would all produce the same effects.Xs However, since the
On this interpretation, the “ proper light” o f each planet is really solar effects differ, so must the powers o f the planets. Therefore each planet must
light received in the interior o f each planet and altered by it. In the final have its own proper light.
analysis, there is only one kind o f light, and it manifests itself in different But then, as if to nullify this argument, Amicus poses a difficulty: different
and unequal w ays.82 This interpretation gains further credibility in the fol­ sublunar effects may occur not only because o f light but also from other
lowing paean to the Sun offered by Mastrius and Bellutus (De coelo, disp. powers o f the planets. Consequently, one cannot be certain that light does
2, qu. 5, 1727, 3:506, col. 2, par. 160): differ in each planet, because the different effects may be caused by other
planetary powers. But if no other power operates except light, then Amicus
From these things it is obvious h o w the Sun is the first measure in the genus o f concedes that the light must be received in a different way in different
lights and the measure o f others and the source o f light. For it exceeds the others planets.86 Here Amicus seems to acknowledge that the “ proper light” o f
and com m unicates light to all. It can be spoken o f as i f it were the o n ly light because each planet is simply sunlight differently adapted by the specific natures o f
by com parison to it the other [lights] can be thought o f as shadow s. each planet. He does not seem to conceive o f a “ different” kind o f light in
each planet. There can be no doubt, however, that Amicus conceived o f
Although Mastrius and Bellutus believed that solar light was the only kind
the Moon as a partially transparent planet capable o f receiving and retaining
ot light, it could take two forms. In effect, they postulated tw o kinds o f solar light in the transparent part and that he considered the remaining
light, pure sunlight and sunlight that is altered within the planet itself. Earlier
planets as totally transparent and capable o f receiving solar light throughout
on, Bartholomew Amicus held a similar opinion, believing that planets had
their bodies.87
some light o f their own in addition to solar light. But Amicus was even
vaguer than Mastnus and Bellutus.
83. In a fourth conclusion. Amicus says that “ Besides a proper light, all planets receive light
etiam propriam luccm habere majorem quam sit albicatio ilia lunae.” Mastnus and Bel­ from the Sun” (Quarta conclusio: omnia astra praeter lumen proprium recipiunt lumen
lutus, De coelo, disp. 2, qu. 5, 1727, 3:506, col. 2, par. 160. a sole). Amicus, De coelo, tract. 6, qu. 5, dubit. 1, art. 3, 1626, 360, col. 2.
50. Ibid. According to Mastrius and Bellutus, someone named Fontana claimed to have 84. Ibid., 359, cols. 1-2.
observed horns o f light on Mars, which resembled those on the Moon. 85. This argument seems to conflict with the fifth argument just described. In that argument,
51. The text for the first reason is: “ Secunda pars quod etiam sint astra ex se ipsis lucida all planets should be lucid in the same way as the Sun, because they are all generically
potest suaden. . . ex hoc: quod omnia alia astra sunt diaphana. ut dicemus. ergo si lucida the same. Nevertheless, the light will vary from planet to planet because the planets differ
non essent, non viderentur, vel saltern non tarn lucida conspicerentur. sicut nec coeli specifically and each will alter the generic light in its own unique way. In the third
cernuntur propter diaphaneitatem;” and the text for the second reason: “ Quia planetae argument, however, Amicus says that if the planets receive only sunlight, they would
. . . apparent diversum coloris quod indicat in luce differre et tamen si a sole tantum all exercise the same effects. But why should this happen? Would their specific differences
illuminarentur deberet lumen in omnibus esse eiusdem rationis." Ibid. not cause each planet to adapt the light to its own specific nature, so that each planet will
52. Mastrius and Bellutus. ibid., insist that the light in each planet is not equally intense. cause different effects?
Since the planets are in different species, equal intensities are not essential (An vero haec 86. Amicus, De coelo, tract. 6, qu. 5, dubit. 1. art. 3. 1626, 358, col. 2.
lux sit aeque in omnibus astris intensa, negative respondendum quia nulla apparet ne- 87. After describing the earth as the most opaque of all bodies. Amicus declares (ibid., art.
cessitas hujus aequalitatis, maxime quia sunt specie distincta). 4, 362, col. 1): “ Ita luna est minus diaphana quam alii planetae. Unde habet in se tarn
412 THE CEL EST IA L REGION CELESTIAL LIGHT 413
N ot all who believed that sunlight was both the ultimate source o f pla­ the former was transformed into the latter. The puzzle might have been
netary self-luminosity and also the source o f ordinary planetary light con­ resolved if Serbellonus had explained how the Sun, as an opaque bodv,
ceived o f planets as wholly or partly transparent. An exception was radiates light to the other planets and fixed stars and how its light is dis­
Sigismundus Serbellonus, who argued that planets and stars possess a source tributed. Does the light o f the Sun lie only or. its surface or deep within?
o f light within themselves that is independent o f the SunHKand also insisted O n the basis of Serbellonus’s discussion, we may assume that whatever
that they are illuminated by the Sun.*9 Arguing from a comm on type o f solution he proposed for the Sun would apply to every other celestial bodv,
analogy favorable to his position, Serbellonus concluded that all stars and each o f which is essentially the same as the Sun.
planets are self-luminous, because they are solid, opaque, and composed o f Few, if any, scholastics took seriously the medieval Avicennan-Macrobian
the same substance. Since they are all alike in these and other vital properties, theory o f self-luminosity, which held that only the Moon received light
he assumes that they belong to the same species. From this, it seemed a from the Sun but that all other planets were self-luminous. The best that
reasonable inference that if the Sun had the property o f light (lux), so also could be said for this theory in the Middle Ages is that Nicole Oresme and
should the planets and stars.90 Since they shared essential properties, we Albert of Saxony thought neither it nor its rival (that all planets receive
may plausibly infer that Serbellonus, who is silent on this, assumed that their light from the Sun) was demonstrable. Although Oresme thought the
the light possessed by each planet and star was identical in its properties to self-luminosity theory more probable, it found little support and only oc­
the light o f the Sun. casional mention.92
Despite the mysterious and often obscure manner in which the “ proper
light” o f the planets is described and defended, it seems likely that the c. The fix e d stars
proper light o f a planet was usually thought o f as sunlight that the planet
had somehow transformed below its surface. M ost o f these scholastics With regard to the planets, most scholastics assumed that solar light was
would also have conceived o f the planets as partly or w holly transparent the single source o f planetary illumination but was manifested in different
bodies capable o f absorbing, retaining, and diffusing sunlight, in a manner ways because o f the diverse natures o f the planets themselves. But what
similar to that described by Albertus Magnus and Albert o f Saxony. about the fixed stars? Did the Sun also illuminate them, or did they produce
Serbellonus was a notable exception, because he assumed solid, opaque their ow n light independently o f the Sun?
planets and stars. He insisted that the light o f the Sun reaches all the planets In turning to this topic we must, at the outset, address a problem o f
and even the fixed stars. Because planets, including the Sun and stars, are terminology. The term planeta (plural: planetae) was always used to signify
opaque and solid bodies, they reflect the Sun’s light and are illuminated. a planet, never a fixed star, whereas the expression Stella f x a (plural: stellae
As opaque bodies, the planets and the fixed stars do not receive solar light fixa e) was the most common expression for a fixed star.9* But three terms
into their depths but only at the surface, where it is reflected. Thus Ser­ caused considerable confusion, namely aster (plural: astra); sidus (plural: sidera
bellonus rejected the idea o f transparent planets that could receive solar light or sydera)-, and Stella (plural: stellae). At some point in the historical evolution
into their interiors.91 The planets were bright because o f reflected light and o f these terms, they seem to have had more or less specific meanings.
also, presumably, because o f their ow n proper light, which, as w e saw, is According to Macrobius (Commentary on the Dream o f Scipio, bk. 1, ch. 14,
probably identical with sunlight. But how could planets produce their own 1952, 147), the term Stella was used for the five planets as well as for those
proper light if the Sun’s light did not penetrate below their opaque surfaces? fixed stars that Ptolemy had not included in his forty-eight constellations.
If Serbellonus retained the two kinds o f light, namely direct sunlight and Thus Stella could be used for five o f the planets (presumably not for the
sunlight transformed into a planet’s proper light, he did not explain how Sun and Moon, however) and for all fixed stars not in a constellation
recognized by Ptolemy. But the terms aster and sidera (sidus, but Macrobius
varias partes, ut macula appareat et lumen non recipiat secundum totam protunditatem
sicut recipiunt alii planetae.” 92. In the late fifteenth or early sixteenth century, Bricot. De celo, bk. 2, i486, 22r, col. 1,
88. “ Stellae et planetae omnes habent lumen ex se.” Serbellonus \De caelo, disp. i, qu. 5, art. who seems to have assumed that the Sun was the source o f light for the planets, also
i], 1663, 2:46, col. 1, par. 3. mentions a challenge to the latter interpretation when he reports the opinion o f those
89. “ Omnia sydera illuminari etiam a sole.” Ibid., col. 2, par. 7. I have interpreted the term who argue that only the Moon receives light from the Sun, whereas the other planets
sydera to be equivalent to stellae et planetae, in note 88. have their own proper light independently o f the Sun. He does not mention Avicenna
90. After declaring that “ Stellae et planetae omnes habent lumen ex se” (see n. 88), Serbellonus or Macrobius. In the seventeenth century, Arriaga thought that it was doubtful that
justifies his assertion: “ quia sunt eiusdem substantiae, soliditatis et opacitatis; magis enim planets (planetae) derived their light from themselves (An planetae habeant a se aliquam
et minus non variant speciem. Ergo non est maior ratio cur lux sit proprietas solis et non lucem plane dubium). Arriaga, De caelo, disp. 1, sec. 6, 1632, 507, col. 2.
aliorum syderum.” Serbellonus, ibid., 2:46, col. 1, par. 3. 93. Occasionally we see an expression such as astra fixa, which appears in Mastrius and
91. “ Dico igitur lumen solis recipi in sola superficie, sive planetarum, sive stellarum fixarum.” Bellutus, De coelo, disp. 2, qu. 2, art. 3, 1727, 3:495, col. 2, par. 78. Some version o f the
Ibid., 2:47, col. 1, par. 10. adjective fixus, fixa. Jixum almost invariably signified a fixed star.
414 THE CELESTIA L REGION CELESTIAL LIGHT 415
actually uses the plural form) were to be applied only to the fixed stars, the sense o f “ fixed stars,” but only o f planets.97 Indeed, the term planeta
with aster indicating a single star and sidera a constellation o f them, such as occurs only once in the question, in the expression “ stelle sive planete que
Aries or Taurus. Although M acrobius’s distinctions were repeated, albeit sunt sub sole scilicet Venus et M ercurius,” which the translator correctly
with some occasional changes and distortions,94 they gradually dissolved, renders as “ the stars or planets which are under the sun, such as Venus or
and the three terms came to be used indifferently for planets and fixed stars. M ercury.” 98 The term Stella is thus a synonym for planeta and turns up in
The term planeta was frequently replaced by one o f our three terms, es­ such expressions as Stella Saturni (the star, or planet, o f Saturn).99 Oresme's
pecially Stella. For example, when Melchior Cornaeus (De coelo, tract. 4, question, which at first glance seems directly pertinent to our present con­
disp. 2, qu. 4, dub. 9, 1657, 512) discussed “ An stellae accipiant lumen a cern about the source o f the light o f the fixed stars, is in fact irrelevant.100
sole,” he was actually asking “ Whether the planets receive light from the This difficulty was already inherent in thirteenth-century discussions about
Sun” and was in no manner concerned with the fixed stars. Here, as in the luminosity o f celestial bodies. Thus Bartholomew the Englishman held
most instances, Stella signifies planet rather than star. Amicus, who usually that the stellae are essentially self-luminous but required supplemental light
used the term Stella for planet, introduces at one point the term sydera and from the Sun. However, in declaring that all stellae except the Moon have
consciously uses it to represent both planets and fixed stars.95 In the course their own proper light, Bartholomew means to include only the planets
o f two successive sentences, Johannes Velcurio describes Jupiter and Mars, under the term stellae.10' For the most part, then, Stella means planet rather
respectively, as sydus Iovis and Stella Martis. And within a few lines, he again than fixed star, usage which continues into the seventeenth century.
speaks ot the lumen syderum as derived from the Sun and also proclaims that Inspection o f medieval discussions about the light sources ot celestial
all the astra are equally lucid. Velcurio used these three terms indifferently bodies reveals that a considerable number o f scholastics discussed the light
for the planets and the term stellae Jixae for the fixed stars.9'1 Although the source o f the planets but that few included the fixed stars. Interest in the
context often determines the meaning o f these terms, occasions arise where light source or sources o f the fixed stars becomes manifest in the late
an author may say something about stellae and leave it to the reader to sixteenth century, when the Coim bra Jesuits declare that the more com­
determine whether he is referring to planets or fixed stars alone, or both. mon opinion o f astronomers was that both the fixed stars and the planets
With this in mind, it appears that few in the Middle Ages inquired about receive their light from the Sun,102 an opinion that was echoed some
the source o f light for the fixed stars, although it may appear they do, seventy years later by George de Rhodes, who declared that “ all the stars,
because o f these terminological problems just described. Thus when Oresme both fixed and errant, seem to have no proper light but are all illuminated
discusses at some length the question “ Consequenter queritur utrum omnes by the light [lux] o f the Sun.” '°- De Rhodes insisted that the fixed stars
stelle habeant lumen a sole vel alique ex se,” which the translator renders were not so far away that the Sun's light could not reach them, as some
as “ Consequently, it is sought whether all the stars receive their light from argued. In a slight modification o f that opinion, Pedro Hurtado de Men­
the Sun [or whether] some stars [produce light] in themselves,” inspection doza also assumed that the sidereal heaven had its light from the Sun but
ot the whole question reveals that Oresme is not speaking o f “ stars,” in allowed that each star might have a very small amount within i t s e l f 04

94. According to Stahl (Macrobius. Commentary on the Dream of Scipio, 1952, 147, n. 41), 97. Oresme, De celo, bk. 2. qu. 11, 1965, 637 (Latin), 638 (English).
Isidore o f Seville and Honorius o f Autun were influenced by Macrobius’s distinctions. 98. Ibid., 64s (Latin). 646 (English).
In an interesting passage, Vincent o f Beauvais. Speculum naturale. bk. is, ch. 16. 1024, 99. Oresme also uses the term astra. as in the expression "Septimo sol lucet ex se, ergo et
col. 1102, claiming to draw upon Isidore o f Seville, says that a Stella is anv single star; alia astra." But astra is simply a synonym for Stella, and both mean planet. Kren translates
sydera are many stars, such as the Pleiades; and astra are great stars, such as Onon and both stelle and aster as “ star” or “ stars." undoubtedly because they both subsume the
Luciter. He goes on to say that “ authors confuse these names and use astra for stellae and term "planet." But in the context o f Oresme’s discussion, they can only mean planet.
stellae for sydera.” Vincent seems to identify all three terms with the fixed stars. Imme­ 100. Indeed, we have already considered Oresme’s important question with respect to the
diately after, however, he declares that “ Stars [stellae] do not have proper light but are Moon and the planets.
said to be illuminated by the Sun, just like the M oon." Here stellae seems to signify the 101. “ Unde omnes stellae habent lumen proprium praeter lunam. Et quamvis stellae ex se
planets. sint luminosae, ad consummationem tamen suae luminositatis recipiunt complementum
95. Because o f its obvious relevance to our discussion. I translate the passage: “ O f stars ab ipso Sole.” Bartholomew the Englishman. De rerum propnetatibus, bk. 8, ch. 33, 1601.
[syderum], some always preserve the same distance between them, as do the 'stars o f the 420.
firmament’ [sydera firmamenti]. Some [however] do not preserve the same distance, as [for 102. For the passage, see note 53 ot this chapter.
example] the planets [planetae]. The stars [sydera] o f the first kind are moved in the same 103. “ Astra omnia, turn fixa. turn errantia, nullam videntur habere lucem propriam, sed ea
way, because if [they were moved] in a different way, they would not preserve the same omnia llluminan a luce solis.” De Rhodes. De coelo, bk. 2, disp. 2, qu. r. sec. 3, 1671,
distance. But it is not so with the other [stars, or sydera, namely the planets] because they 284, col. 2.
must necessarily be moved with different motions, because otherwise the difference of 104. “ Coelum svdereum nullam aut exiguam habere nativam lucem.” And some lines below,
distance and nearness could not be caused." Amicus, De caelo, tract. 5, qu. 7, dubit. 2, he declares; “ Non propterea nego aliquid nativae lucis sydenbus, sed lllud dico esse
art. 1, 1626, 338, col. r. perexiguum coilatum cum lumine quod a sole mutuantur. ” Hurtado de Mendoza. De
96. See Velcurio, Physics, 1.SS4, 7b. col. 2 and 77, col. 1. coelo, disp. 2, sec. 5, 1615, 375, col. 1, par. 64.
416 THE CELESTIAL REGION CELESTIAL LIGHT 417
He based his argument on Genesis 1 .14-15, where, after God had already body, the Sun, Aversa assumes that every fixed star is transparent and
created the firmament, he is said to have created the luminaries in the possesses its own source o f light.
firmament to provide light for it. Thus the firmament is seen only by Giovanni Baptista Riccioli considered the opinions o f Galileo, Kepler,
virtue o f the luminaries, or the Sun and Moon. If it had its ow n native and Descartes, all o f whom assumed that the fixed stars possessed their own
light, it would have been seen before the creation o f the luminaries. light and received none from the Sun, as far more probable than any other
Nevertheless, the stars may have a very small amount o f light, which is, interpretation. In addition to his acceptance o f their reasons, Riccioli thought
however, insufficient to make them visible to us without the Sun. Thomas it was more appropriate “ to the majesty o f the Divine Creator that there
Com pton-Carleton was less tentative in his attribution o f some light to not be a single light for the stars but that a multitude should light in the
the fixed stars when he said that “ neither the Moon, nor the other planets, manner o f the Sun. N or do they [the fixed stars] require another source o f
nor even the fixed stars receive all their light from the Sun,” 10510 6 thus light other than God, the Father o f all lights.” T o strengthen his argument,
implying that the fixed stars receive part o f their light from some other Riccioli quotes from Baruch 3.34-35, where it is said that “ The stars shone
source or from themselves. ,°'> at their appointed stations and rejoiced; he [God] called them and they
But scholastic authors were hardly ot one mind, and some were con­ answered, ‘We are here!’ Joyfully they shone for their M aker.” ,ov The stars
vinced that the fixed stars were self-luminous. In 1627, a few years before were thus capable o f shining with their own light and had no need o f the
Galileo likened tne fixed stars to “ so many Suns” (Dialogue, Third Day Sun.
[Drake], 1962, 327), thus implying their self-luminosity, Aversa left no Johannes Poncius, and perhaps Roderigo de Arriaga, also joined the ranks
doubt as to his interpretation: the fixed stars probably are not illuminated o f those scholastics who regarded the fixed stars as essentially self-luminous
by the Sun but are selt-luminous.10710 8The planets are related to the Sun, bodies. Poncius couched his statement in a manner analogous to the way
which lies in their midst and with respect to which they move. Therefore in which he identified the source o f planetary light. Just as the Sun is
it is proper that they should receive their light from the Sun. But the principally (principaliter) the source o f planetary light, so also do “ the fixed
fixed stars have no such relationship to the Sun and derive no light from stars have their light principally from themselves.” '" Poncius describes this
it; rather they have their own proper light. Moreover, by contrast to the opinion as “ more com m on.” " 2 He could find no experience that would
opaque bodies o f the planets, which receive light only on their surfaces, lead us to believe that the fixed stars receive their light from the Sun. But
the fixed stars are transparent bodies that are suffused with light through­ he invokes one experience that indicates that the fixed stars are self-
out their depth. '0* Thus the transparency that Albertus Magnus, Albert luminous, namely the fact that they are visible at noon from the deepest
o f Saxony, and a number o f early modern scholastics attributed to the w e lls ."3
planets, Aversa assigns only to the fixed stars. And whereas Albertus and
Albert filled the transparent planetary bodies with light from another
109. New English Bible, 1976, p. 179 (of the Apocrypha).
n o. Indeed, Riccioli (Almagestum novum, pars post., bk. 6, ch. 2, 1651, 395, col. 2) thought
105. “ Nec luna, aut alij planetae, nec stellae etiam fixae omnem lucem suam accipiunt a sole.” it natural to ask which o f these two lights was stronger, that o f the Sun or that o f
Compton-Carleton, De coelo, disp. 2, sec. 3, 1649, 403, col. 1. the fixed stars. He concluded that they might have light o f equal strength (ibid., 396,
106. Compton-Carleton offers no further elaboration. col. 1).
107. In the late sixteenth century, Giordano Bruno (1548-1600), and perhaps even Nicholas i n . “ Stellae fixae habent lucem suam principaliter a seipsis.” Poncius, De coelo, disp. 22,
o f Cusa (1401-1464) in the fifteenth century, had assumed that all stars were Suns. See qu. 8, 1672, 625, col. 1, par. 74.
Dick, 1982. 108 for Bruno, and ibid., 40, tor Cusa, who assumed that all celestial bodies 112. “ Haec est communior.” Ibid.
were like the earth, which he characterized as a star that was self-luminous. By impli­ 113. “ Et non constat ulla experientia quod stellae illae mutuent a sole suam lucem; imo potius
cation, then, all stars should be self-luminous. suffragatur experientia opposito, nam in ipso meridie ex altissimis puteis stellae fixae
108. “ Deinde haec omnia nullatenus procedunt de stellis fixis atque adeo nullum prorsus videri possunt." Ibid. It is on the basis o f this same argument that Arriaga may be
indicium suppetit ut lumen suum dicantur a sole recipere. Et quidem verisimile videri classified among those who believed that the fixed stars are self-luminous. He thought
poterit bene quidem planetas, non autem Stellas fixas illuminari a sole quia planetae it probable that “ the light o f these stars [stellae] is independent o f the Sun.” Although
agnoscunt pro suo principe solem et in medio ilium continent et in suo motu habent the passage, which follows, may also be interpreted to apply to the superior planets
respectum ad solem bene ergo putari debent ab illo et per ilium lucere. At stellae fixae instead o f the fixed stars - the term stellae is equally applicable to both - the fixed stars
per se propriam agunt aciem et non habent huiusmodi respectum ad solem. Censeri seems more appropriate in light o f Arriaga’s example, which is the same as that o f
ergo potius debent non illius beneficio sed propria virtute lucere. Et iuxta hoc planetae Poncius. Here is the relevant passage; “ An planetae habeant a se aiiquam lucem plane
quidem did debebunt corpora opaca atque in solo externo ambitu lumine illinita. Stellae dubium. Probabile est habere quia si vera est sententia docens etiam in meridie videri
autem fixae esse corpora perspicua et in tota sua mole ac profunditate eodem suo lumine ex profundis puteis Stellas, cum ipsae sint supra solem et tunc a sole non respiciantur ne
perfusa. Iuxta id quod dicebamus quaestione superiori sect. 2. Et hoc etiam iuuare potest illuminentur (suppono enim solem esse opacum a tergo, ut possit ad nos melius lucem
ad rationem reddcndam cur maxime stellae fixae scintillent et sol non solum scintillare mittere), necessarium est ut ilia lux stellarum sit independens a sole.” Arriaga, De caelo,
sed et veluti ebullire cernatur.” Aversa, De caelo, qu. 35, sec. 3, 1627, 173, col. 2-174, disp. 1, sec. 6, 1632, 507, col. 2. For Compton-Carleton’s use o f the well argument,
col. 1. see note 124.
418 THE C EL EST IA L REGION CELEST IAL LIGHT 419
By the time Poncius published his relevant work, in 1642 and 1643,1,4 it observations associated primarily with the names o f Copernicus, Brahe,
is likely that the assumption o f the fixed stars as self-luminous rather than and Galileo.
illuminated by the Sun was indeed “ more com m on.” If it was not more
common among scholastic authors, it was nevertheless widely accepted.
H ow was it that the Sun, which had previously been thought to be the IV . Is the lig h t o f the stars and planets
primary light o f the world on which other bodies depended for their illu­ o f the sam e species?
mination, came to be perceived as no more light-giving than a fixed star
and perhaps even less so? From the time o f Richard o f Middleton in the thirteenth century, those
The explanation may he in the newly developed ideas about a vastly who attributed a certain degree o f weak self-luminosity to the planets and
expanded universe that had its roots in the Copernican theory. Because who also assumed the Sun as the primary source o f planetary light had
no stellar parallax could be detected from the earth's orbital motion, an seemingly recognized at least two different species o f light in the heavens.12,0
enormous spatial gap had to be assumed between Saturn and the fixed Although this might have seemed a logical consequence based on the dis­
s t a r s . I n the Copernican scheme, the universe was of enormous size, tinction o f two different lights, it was probably a minority opinion.
perhaps even unmeasurable, and the fixed stars were a vast distance from As with almost all scholastic issues, arguments for another interpretation
the Sun. But it was not merely a matter o f distance. Tycho Brahe, an could usually be formulated with some plausibility within the Aristotelian
opponent o f the Copernican system, showed that if the Copernican scheme system. In this case, however, it was probably a strong desire to treat nature
were true, a third-magnitude star would have a diameter 200 times greater in the simplest terms and therefore to subsume all light under a single
than that o f the Sun, which he thought absurd."" It would seem an odd species. Thomas Compton-Carleton, for example, insisted that, with the
universe in which a relatively small Sun could illuminate huge stars so exception o f the light in the empyrean heaven, which differs from the light
very tar a w a y ."’ o f the physical world, there is only one species o f light, not only in the
But scholastics were not Copermcans. and those who followed Tycho heavens but also in the sublunar w orld .1' 1 The light for all these bodies is
may have known that he had not only assumed a smaller distance o f the derived ultimately from the Sun. But what o f the claim that the planets
fixed stars than was traditional but also reduced their size ."' What reason have their own light, independently o f the Sun? Does this not make it differ
would scholastic natural philosophers have had to assume that the fixed in species from the light o f the Sun? In a strange reply, Compton-Carleton
stars were self-luminous and independent o f the Sun? Perhaps Galileo’s explains that the natural light o f the planet is intensified and supplemented
analysis of his telescopic observations impressed them. After all, Galileo, by solar light, which could not occur if they were really distinct in species, 122
and later Kepler, had declared that the stars were themselves Su n s."<; As a point that Amicus, who also argued that celestial light does not differ
Suns, the fixed stars would provide their own light. specifically, elaborated some years before when he remarked that an internal
No compelling argument or overwhelm ing piece o f evidence promoted light could be intensified by an external light only if they were o f the same
and encouraged the partial scholastic acceptance o f self-luminous fixed stars. species. T o be intensified to a more intense degree implies that the thing
Indeed, during the Middle Ages the question about the light source o f the intensified and the thing intensifying are the same kind o f thing - that is,
fixed stars was rarely, if ever, raised. Those to whom the question did occur in the same species.1-’
probably assumed that the Sun was the source o f all stellar light. N ot until 120. For example. Richard o f Middleton emphatically distinguished two species ot light. See
the late sixteenth and the seventeenth century did the issue acquire a measure note 40. this chapter.
121. The Coimbra Jesuits (De coelo, bk. 2, ch. 7. qu. 9, art. 2, 1598, 301) adopted a similar
of prominence, and this solely because of revolutionarv new concepts and
attitude, holding that all light produced by the forces o f nature belonged to the same
species. After expressing some doubts, they even included the lights associated with
“ the glorious bodies. ” whose light was said to be as lucid as that o f the Sun and therefore
1 14. Although I have used the 1072 edition, the first edition of 1 volumes appeared in 1042- in the same species.
1643 (see Lohr. 19NX, 302). 122. “ Quoad lucent autem aliorum coelorum probatur earn specie non distingui. Recipiunt
1 is. See Van Helden, 19X3, 40-48. quippe a sole lucem. sicut aer et res omnes sublunares. Ergo non est cur lux ilia sit
110. Ibid., 51-32. diversae ratioms .1 luce solis. Dices primo secundum nos singuli caeli et astra habent
1 1?• Otto von Guericke emphasized the enormous distance between the Sun and the nearest aliquid lucis ex se. ergo 1II.1 saltern erit diversae speciei secundum diversitatem substantiae
fixed star as a major factor in denying that sunlight could light the stars, even though a qua oritur. Negatur tamen consequentia ilia enim lux intenditur a luce solis quod tamen
it lit the planets. Guericke, 1072, 229, col. 1. fieri non posset si esset specie adaequate distincta.” Compton-Carleton, De coelo, disp.
118. Van Helden, 1985, 51. 2, sec. 3, 1649. 4^3. col. 2.
119. Guericke (1672, 230, col. 1) cites both Galileo and Kepler as among those who identified 123. “ Secundo quia lumen intrinsecum intenditur ab extnnseco, ergo sunt eiusdem speciei.
the stars as Suns. For the former he cites the Sitiercus mitiaus G010), for the latter, nam intensio non tit nisi per gradus qualitans unius speciei.” Amicus, De coelo, tract. f>,
Disscrtatio cum Xuncio sidereo ( i O i o ). qu. 5, dubit. 2. 1020. 305, col. 1.
420 THE CELESTIA L REGIO N CELESTIAL LIGHT 421

But what about the seeming variety o f lights in the heavens, namely the claims made in the Copernican and Tychonic systems and knew about
light o f the Sun, the planets, and the fixed stars: Are they not different? In Galileo’s telescopic discoveries. Despite their commitment to Aristotelian
denying any differences, Com pton-Carleton points to the rainbow, which, cosm ology on traditional and religious grounds, scholastics sometimes
although it contains different colors, is produced by one kind o f light. adopted ideas from the new science and incorporated them into their own
Differences in light are only apparent and arise from differences in the cosmic scheme without great difficulty. As a consequence, scholastic cos­
properties o f bodies, especially density and transparency. M oreover, the m ology encompassed both traditional and new elements, even though the
light o f the Sun does not destroy the light o f the planets and fixed stars, latter were rather poorly integrated into the Aristotelian system. It is difficult
although it may seem to do so, because their light is not seen during the to imagine how a hierarchically based Aristotelian cosm ology could easily
day. But if you descend into a deep well, where the Sun’s light cannot reconcile a planetary system which had the Sun, the fourth planet from the
reach, you will perceive the stars clearly, as at night. What makes the stars earth, as the primary source o f all planetary light and could also assume
and planets invisible during the day is the principle that if a sensible thing, self-luminosity for the fixed stars. The relationship between the fixed stars
say the Sun’s light, exceeds by a great amount a lesser sensible thing, say and the Sun was thus complex. T o add to the difficulties, Galileo had called
the light o f individual planets and fixed stars, the greater will drastically the fixed stars Suns. Did this imply that the Sun was therefore a fixed star?
interfere with the smaller.124 For Copernicans, who assumed an immobile Sun, the Sun could indeed be
conceived as a fixed star, but not for Aristotelians, who assumed a Sun that
moved around a stationary earth. For the most part, scholastics who thought
the fixed stars self-luminous avoided such questions and their implications.
V . C elestial lig h t as a m ix o f o ld and n ew They coped with questions about light in the usual ad hoc manner. Although
Despite their contributions, however, most scholastic arguments about the light was perhaps the most important and spectacular attribute o f the heav­
sources o f light were, as we have seen, not technical but general, analogical, ens,125 there were other properties and powers that marked out the heavens
and scriptural. Raphael Aversa, for example, did not argue that the fixed as special. It is now time to describe them.
stars were Suns. Indeed, he assumed they were totally different: whereas 125. Among the visible qualities, Aversa, De caelo, qu. 34, sec. 3, 1627, 133, col. 2, called it
the Sun was an opaque body, the fixed stars were transparent and self- “ the most noted.”
luminous. For thinkers who followed a centuries-long tradition about the
hierarchy o f the heavens, it may well have seemed perfectly plausible for
the higher, and therefore nobler, fixed stars to have their own light rather
than be dependent on light from a planet so much closer to the earth and
therefore presumably less perfect. But Aversa joined Galileo in the belief
that the fixed stars were self-luminous and not lighted by the Sun. In this
decision, he was probably influenced by the controversies and debates con­
cerning the new concepts and observations that had emerged since the end
o f the sixteenth century.
What applies to the fixed stars is equally applicable to the entire range o f
problems about celestial light and its manifestation in all celestial bodies.
Many, though by no means all, scholastic natural philosophers in the sev­
enteenth century mention the names o f Copernicus, Brahe, Galileo, and
Kepler. Directly or indirectly, they were aware o f the new and dramatic

124. “ Dices secundo apparet alterius quasi rationis lux in stellis ac veluti fulva. Contra etiam
in iride apparet lux diverse rationis et tamen non est, sed prorsus eiusdem. Hoc ergo
solum provenit ex diversa dispositione corporis in quo recipitur secundum diversam
temperiem, densitatis. diaphaneitatis, etc. Nec lux solis destruit lucem planetarum aut
stellarum firmamenti quod interdiu non appareant, sed hoc ex eo provenit: quod maius
sensibile, praesertim si vaide magnus sit excessus, impediat minus. Unde si quis medio
die descenderet in protiindum puteum quo lux solis non pertingeret aeque clare. ut aiunt,
perciperet Stellas, ac nocte. ” Compton-Carleton, De coelo, disp. 2, sec. 3, 1649, 403,
col. 2-404, col. 1.
PROPERTIES OF CELESTIAL BODIES 423
1 7
origin or composition were drawn from the elements.” 4 A t approximately
the same time, John o f Sacrobosco declared in his famous Sphere that
“ Around the elementary region revolves with continuous circular motion
The properties and qualities the ethereal, which is lucid and immune from all variation in its immutable
essence. And it is called ‘Fifth Essence’ by the philosophers.” 5 Elaborating
on these few sentences, Christopher Clavius reveals that the ether was
o f celestial bodies, and the understood in much the same w ay nearly four centuries later.6 He distin­
guished five major properties, the first o f which is that the ethereal region
dimensions o f the world encloses the elementary region as its container and is therefore its place. As
the place o f the elementary region, philosophers consider the ethereal zone
more excellent, because it is removed from the incessantly changing region
o f the elements and also because it exists among the divine movers o f the
orbs that enjoy the best life. The second o f its properties is light, which is
“ One may find it surprising,” wrote Friedrich Solmsen in his informative much nobler than that o f the elements. As the third property o f the celestial
study o f Aristotle’s physical system o f the world, “ that Aristotle does not ether, Clavius mentions its capacity for avoiding change: it cannot be altered,
say more about the nature o f the heavenly bodies.” ' O n this subject, A r­ or diminished, or increased, or generated or corrupted, all o f which attri­
istotle’s commentators had no choice but to find their ow n way. butes are the opposite o f those in the four elements. Its fourth property is
its continuous circular motion, which is the cause o f continuous generation
and corruption and stands in contrast to the natural rectilinear motion o f
the inferior, terrestrial region, which is not perpetual but always comes to
an end. As its fifth and final property, Clavius observes that philosophers
I. The celestial ether
call the ethereal region a fifth essence. For centuries, scholastics had thought
Because planets, stars, and orbs were assumed to be constituted o f a special o f the celestial ether in virtually the same terms.
celestial ether, I have had occasion to mention it at different points in this U p to this point, we have had occasion to consider the celestial ether in
study. It is now time for a more systematic examination o f this extraordinary a number o f different contexts. In Chapter 10, w e examined its incorruptible
substance.12 nature, and in Chapter 12 considered whether or not it was perceived as
With the notable exception o f Robert Grosseteste, and perhaps a few matter (for the most part, it was) and whether that incorruptible celestial
others during the Middle Ages, the celestial region was assumed to be substance was a composite o f matter and form, and whether celestial matter
composed o f ether, rather than o f fire or some combination o f the four was different from or identical to corruptible terrestrial matter. And in
elements.3 In his popular encyclopedia, Bartholom ew the Englishman de­ Chapter 14, we pondered the fundamental question as to the hardness or
scribed Aristotle’s ether as “ something beyond the lunar globe that is o f a fluidity o f the ethereal orbs. In these primarily metaphysical discussions,
separate nature from the nature o f the inferior elements. Thus the ether is the level o f discourse was mostly abstract and general. We must now ap­
neither heavy nor light, neither rare nor dense, nor is it divisible by the proach the ether as a substance comprised o f a large number o f seemingly
penetration o f another body. N o corruption or alteration, universally or different bodies: planets, stars, and orbs. Indeed, one may even ask whether
particularly, affects the ethereal nature, which would happen to it if its the waters above the firmament, which form the crystalline or ninth orb,
could be formed o f the same matter as the firmament, which separates the
waters forming the crystalline orb from the waters below .7 If the ether is
1. Solmsen, i960, 316, n. 50.
2. For Solmsen’s description o f Aristotle’s conception o f the ether, or “ first body,” as the
latter called it, and the role it played in his system, see ibid., 287—309. 4. “ Quicquid enim supra lunarem globum est. naturae est separatae a natura inferiorum
3. Grosseteste, Degeneratione stellarum, 1912, 33, says that “ a star does not possess the nature elementorum. Unde aether neque est grave neque est leve; neque rarum, neque densum;
o f a fifth essence” (Stella autem non est de natura quintae essentiae). Indeed he also argues neque per alterius corporis penetrationem divisible. Naturam enim aetheream nulla in-
(ibid.) that stars are not only composites o f matter and form but also composed o f elements, greditur corruptio vel alterado universaliter vel particulariter, quod ei accideret si ex de­
by which he clearly means elements that do not differ from our terrestrial elements. mentis compositionem aut originem contraxisset. ” Bartholomew the Englishman, De rerum
Grosseteste was but following the earlier Platonic tradition characteristic o f the early Middle proprietatibus, bk. 8, ch. 5 (“ De aethere” ), 1601, 381.
Ages, which became popular again in the sixteenth and seventeenth centuries when the $. Sacrobosco [Sphere, ch. 1], 1949, 119.
observation o f seemingly real changes in the sky undermined the notion o f celestial in­ 6. What Follows appears in Clavius [Sphere, ch. 1], Opera, 1611, 3:20.
corruptibility and consequently also o f the ether that was allegedly incorruptible. 7. At first glance, this question seems to cry out For a response, but to my knowledge it was

422
424 THE CELESTIAL REGIO N PROPERTIES OF CELESTIAL BO DIES 425

a single substance, w hy do many o f the celestial bodies appear to differ to identify it as a problem and offered little help in its resolution. Scholastic
from one another? natural philosophers devoted no questions to it. And yet the problem was
If the medieval followers o f Aristotle had taken his statements about the implicitly acknowledged, because it was discussed in an indirect manner.
ether in De caelo literally, or as Bartholom ew the Englishman and Clavius Celestial properties have tw o distinct aspects, visible and invisible. The first
understood them, they would have been compelled to conclude that all o f these concerns such visually detectable variations as brightness and color
celestial bodies and the ether as a w hole possessed the same properties. For, among stars and planets or the radical differences between the Sun and the
on the assumption o f ethereal homogeneity, it follows that the ether and Moon; the second applies to latent and invisible properties and powers that
all the celestial bodies within it are identical in appearance and power and were assumed to inhere in some sense in each celestial body and which were
in all distinguishable qualities. And yet Aristotle him self cast doubt upon thought to produce changes in the terrestrial region. Although visible and
this interpretation when, in his Meteorologica (i-3.340b.6-10), he indicated invisible celestial properties were rarely, if at all, distinguished, I shall devote
that the ether was not uniform in quality, especially in those parts bordering a separate chapter (Ch. 19) to the influence o f the celestial region on the
on the terrestrial region. Gross observation, moreover, made it apparent terrestrial, which is largely concerned with latent properties and influences.
that the celestial bodies did indeed differ: the Sun from the M oon, and the In the broadest sense, any scholastic natural philosopher who gazed sky­
tw o latter bodies from the other five planets, which, in turn, seemed to ward on any clear evening could see a panorama o f wandering planets and
vary from each other. For a very long time, astrologers had forecast their fixed stars, between which there were vast tracts o f darkness. As we have
predictions and assessments on the basis o f assumed differences in the pow ­ already seen, Aristotle had assumed that this entire celestial region is filled
ers and appearances o f the planets and stars. Aristotle him self recognized w ith a fifth element, or incorruptible, pure, transparent ether. The ether
such differences. The author o f the De proprietatibus elementorum, falsely exists in the form o f gigantic but invisible spheres, within which are embed­
attributed to Aristotle during the M iddle Ages, insisted that the M oon and ded the visible stars and planets. As the spheres turn with circular motion,
other planets diverged from the Sun and that the planetary orbs differed they carry around all the visible celestial bodies. But the concept o f a pure,
from the planets they carried, because the former, although they receive homogeneous, transparent ether posed a monumentally difficult problem,
light, are not illuminated as are the planets, an indication that their substances one that Aristotle ignored but Alexander o f Aphrodisias did not, as we
varied. Indeed, the anonymous author was prepared to distinguish three learn from Simplicius, who reports that Alexander asked how celestial dif­
distinct celestial substances.8 H ow, then, was this apparent conflict between ferences could occur in the simple celestial ether. If celestial bodies differed
Aristotle’s theory o f a uniform, homogeneous, and incorruptible ether rec­ in density or rarity or with respect to color and other properties, how could
onciled with observed differences among the planets, stars, and even the such differences occur in an element in which no changes were thought
orbs themselves? possible because no contrary properties could exist therein?9
We must first recognize that what I have characterized as an “ apparent Alexander’s response, with which Simplicius concurred, was destined to
conflict” was rarely made explicit